Mineralogical-Chemical Alteration and Origin of Ignimbritic Stones Used in the Old Cathedral of Nostra Signora Di Castro (Sardinia, Italy)

Studies in Conservation

ISSN: 0039-3630 (Print) 2047-0584 (Online) Journal homepage: https://www.tandfonline.com/loi/ysic20

Mineralogical-chemical Alteration and Origin of Ignimbritic Stones Used in the Old Cathedral of Nostra Signora di Castro (Sardinia, Italy)

Stefano Columbu, Anna Gioncada, Marco Lezzerini & Fabio Sitzia

To cite this article: Stefano Columbu, Anna Gioncada, Marco Lezzerini & Fabio Sitzia (2019) Mineralogical-chemical Alteration and Origin of Ignimbritic Stones Used in the Old Cathedral of Nostra Signora di Castro (Sardinia, Italy), Studies in Conservation, 64:7, 397-422, DOI: 10.1080/00393630.2018.1565016 To link to this article: https://doi.org/10.1080/00393630.2018.1565016

Published online: 18 Mar 2019.

Submit your article to this journal

Article views: 83

View related articles

View Crossmark data

Citing articles: 3 View citing articles

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=ysic20 STUDIES IN CONSERVATION 2019, VOL. 64, NO. 7, 397–422 https://doi.org/10.1080/00393630.2018.1565016

ORIGINAL RESEARCH OR TREATMENT PAPER IN SARDINIA Mineralogical-chemical Alteration and Origin of Ignimbritic Stones Used in the Old Cathedral of Nostra Signora di Castro (Sardinia, Italy) Stefano Columbu1, Anna Gioncada 2, Marco Lezzerini2 and Fabio Sitzia1 1Department of Chemical and Geological Sciences, University of Cagliari, Cittadella Universitaria di Monserrato, Cagliari, Italy; 2Department of Earth Sciences, University of Pisa, Pisa, Italy

ABSTRACT ARTICLE HISTORY The pyroclastic rocks belonging to the Late Eocene-Miocene volcanic activity that occurred in Received August 2018 Sardinia between 38 and 15 Ma ago were widely used as construction materials in several Accepted December 2018 Romanesque churches of the easternmost Logudoro area, as well as in large parts of the KEYWORDS Sardinia territory. In this work, the ancient Cathedral of Nostra Signora di Castro (twelfth Provenance; XRF; XRD; SEM- century) was taken as a representative case study. There is no historical or archaeological EDX; petrographic study; evidence of ancient quarries. Based on the geochemical, petrographic, and volcanological medieval church; Logudoro; 2 data on several samples from an extensive field area (approximately 150 km ), a Romanesque geographical zoning of the volcanics has been recognised. In the Oschiri sector, there are three different sub-zones, which can be identified with different volcanic rocks: less fractionated rocks (Differentation Index ∼70–78); intermediately fractionated rocks (D.I. ∼76– 79); and more fractionated rocks (D.I. ∼77–82). To identify the origin of the ignimbrite rocks of the Church of Nostra Signora di Castro, two statistical methods were used: stepwise linear discriminant and canonical analysis. Moreover, to define the geochemical transformation processes induced by the alteration, a comparative study of concentrations of major and trace elements measured by XRF and SEM-EDX analyses on the surface portion and the innermost areas of the stone was made.

Introduction and aims the northern side abuts a porch, built later. The cathe- The Church of Nostra Signora di Castro was built in the dral became the prototype of several other churches territory of Oschiri on the Logudoro area (northern Sar- built in the territory of Oschiri and dating before dinia; Figures 1 and 2(a, b)). It was consecrated in the 1174: S. Demetrio, located inside the Oschiri village year 1174, and it was once the cathedral of the (Figure 2(c, d)) and already consecrated in 1168; diocese of Castro (Spano 1858) merged in 1503 with Santa Maria di Otti, located inside the ancient village the diocese of Ottana and Bisarcio. The Church of of Otti (before Octi, Figure 2(e, f)); and S. Maria di Nostra Signora di Castro is one of the smallest Sardinian Tula, already existing in 1176. Romanesque cathedrals, but it represents one the most The Church of Nostra Signora di Castro (Figure 2(a, beautiful Romanesque churches located in rural areas. b)) was built with diﬀerent kinds of local ignimbritic It is located on the site of the ancient Roman fortiﬁed rocks of Late Eocene-Miocene volcanic phase (Becca- military centre of Lugudonec, mentioned in the Itinerar- luva et al. 2011; Lustrino, Duggen, and Rosenberg ium Antonini. The diocese had a populated and vital ter- 2011), whose products occur on vast outcrops ritory during the twelfth century, but already in 1220 (Figures 1 and 3) on several sectors of Sardinia the diocesan appears to be impoverished. The docu- (Figure 1). The ignimbrites were frequently used mentary sources highlight that by the year 1420 the together with other rocks and ancient mortars to site was abandoned, although the Church of Nostra build several historical monuments in the Middle Signora di Castro still remained in use for a long time, Ages (Figure 2) as well as in the other ancient periods least every once in a while (Serra and Coroneo 2004). (Coroneo 1993; Cruciani et al. 2010; Verdiani and The modest size of the Church of Nostra Signora di Columbu 2010; Columbu and Verdiani 2014; Antonelli, Castro seems to be unusual for a cathedral, especially if Columbu, De Vos Raaijmakers, et al. 2014; Antonelli, compared to other coeval cathedrals in the same Giu- Columbu, Lezzerini, et al. 2014; Columbu et al. 2014; dicato of Torres. However, its historical-cultural impor- Chiarelli et al. 2015; Di Benedetto et al. 2015; Miriello tance is evident in technical and formal aspects such et al. 2015; Columbu, Sitzia, and Verdiani 2015; as the care in the cutting and placing of the ashlars Columbu, Cruciani, et al. 2015; Columbu and Garau in the wall surface (Serra and Coroneo 2004). Along 2017; Columbu, Lisci, Sitzia, and Buccellato, et al.

CONTACT Stefano Columbu [email protected] Department of Chemical and Geological Sciences, University of Cagliari, Cittadella Universitaria di Monserrato, Va Trentino 51, Cagliari, CA 09134 Italy © The International Institute for Conservation of Historic and Artistic Works 2019 398 S. COLUMBU ET AL.

Figure 1. Schematic geological map of Sardinia with the local- isation of the Oschiri village and sector reported on the sketch map of Figure 3. Legend of patterns and colours refer to lithol- ogies: white = recent alluvial sediments; light grey = Oligo- Figure 2. Old Cathedral of Nostra Signora di Castro (Oschiri) Miocene volcanics including the Oschiri ignimbrites; dark and other important medieval Romanesque churches: (a, b) grey = Plio-Pleistocene volcanics; stippled grey = Miocene main facade and apse of Castro Church, respectively; (c, d) marine sediments; grey crosses = Paleozoic crystalline base- Santa Maria di Otti Church (south of Castro Church); (e, f) San Demetrio Church, located within the cemetery of Oschiri ment and Mesozoic formations; and red continuous and ’ dashed lines = faults. village; and (g, h) Basilica of Sant Antioco di Bisarcio, located in the ﬁeld of Ozieri.

2017; Columbu, Sitzia, and Ennas 2017; Raneri et al. 2018; Columbu, Antonelli, and Sitzia 2018; Columbu, papers, the stone decay has been investigated, Piras, et al. 2018; Columbu, Palomba, et al. 2018; especially studying sandstone, limestone, or other Columbu, Lisci, Sitzia, Lorenzetti, et al. 2018; Columbu, rocks (Rodriguez-Navarro et al. 1997; Sabatini et al. Garau, and Lugliè 2018; Manzoni et al. 2019). 2000; Bargossi, Felli, and Guerriri 2002; Franzini et al. The ignimbritic rocks probably belong to the local 2007; Benavente, Cultrone, and Gómez-Heras 2008; outcrops, but considering the geochemical variety of Sebastian et al. 2008; Wangler and Scherer 2008; volcanics, it is not easy to find their exact origin. More- Leoni et al. 2010; Gioncada et al. 2011; López-Doncel over, no evidence exists of ancient quarries and it is et al. 2013; Lezzerini et al. 2016; Lezzerini et al. 2018; probable that there were contemporaneous different Ramacciotti et al. 2018), but research on Sardinian pyr- procurement points of geomaterials in the field, or oclastic rocks is still scarce, despite their wide use in the use of stone boulders. ancient times (Coroneo 1993). The volcanic rocks used to build the Church of The main aim of this research was to determine the Nostra Signora di Castro, especially when pyroclastic minero-petrographic and geochemical characteristics of facies were used, show frequently a physical decay volcanic rocks used in the Church of Nostra Signora di and a mineralogical-chemical alteration. In several Castro, in order to identify the origin of the geomaterials, MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 399

Figure 3. Geological sketch map of Oschiri-Chilivani/Tula sector where are located the medieval churches of Castro Cathedral, Santa Maria di Otti, and San Demetrio churches, and Basilica Sant’Antioco di Bisarcio (Ozieri territory). The geological map has been set according to geopetrographic surveys performed in this work, according to Beccaluva et al. (1985), Carmignani et al. (1996), and Lecca et al. (1997). providing geological samples for any restoration inter- archaeological historical point of view (to know the vention on the monument and to deepen knowledge ancient trade of building materials in historical times), of the ancient trade of raw materials in the Middle Ages. but also to intervene in the restoration of monuments A detailed petrographic and geochemical compari- with the substitution, or partial integration, of the son of ignimbritic rocks from the Church of Nostra degraded decorative elements or ashlars of monuments. Signora di Castro and the volcanic outcrops belonging to the Oschiri and Chilivani/Tula areas around the Church was made, using the geochemical XRF data col- lected by Columbu (2017). The purpose of the work Material and methods was also to define the physical and chemical factors A sampling of stones from the Church of Nostra in the decay processes of these ignimbrites. To study Signora di Castro refers exclusively to the building the chemical alteration of the samples from the material taken from the exterior façades. Seventy-five Church of Nostra Signora di Castro, a comparative samples come from ashlars of the ancient building study of concentrations of major and trace elements and they represent all the volcanic lithotypes as for measured by XRF analysis in the surface portion (A+) orientation (N-S-E-W), height above the surrounding − and in the innermost of the stone (A ) was made. countryside, and different alteration degrees. As little The study of the chemical-physical alteration of geo- material as possible in the form of splinters was materials used in cultural heritage allows first of all an removed, still attaining our dual goal of sampling the understanding of the causes and mechanisms of innermost part (A−) and the outermost one (A+), as decay processes that occurred in the monument. Fur- well as having enough material for the required petro- thermore, the definition of the origin of geomaterials graphic and chemical analyses. A− and A+ specimens is important and significant not only from an were analysed only for a total of 29 samples out of 79. 400 S. COLUMBU ET AL.

Two hundred and twenty rock samples from the Berchidda Basin (Oggiano, Pasci, and Funedda 1995), Oschiri area and nearby outcrops (previously studied in Eastern Logudoro (Figure 3). The basin is an by Columbu (2017)) were used to define the geological offshoot of an Oligo-Miocene complex tectonic struc- provenance of the ignimbrites used for building the ture, trending N-S, known as the ‘Fossa tettonica Church of Nostra Signora di Castro. In the same way sarda’ (Vardabasso and Atzeni 1962)(Figure 1)or‘Rift as the samples from the Church, A− and A+ specimens of Sardinia’ (Cherchi and Montadert 1982). This latter were analysed only for a total of 27 samples out of 220. is a large tectonic low (Coulon 1977; Cherchi and Mon- Uncovered thin sections, oriented parallel and tadert 1982; Dostal, Coulon, and Dupuy 1982; Advokaat orthogonal to the flow planes (i.e. magmatic foliation) et al. 2014), where, as in other areas of Sardinia, intense of the volcanic rocks, were prepared for optical micro- igneous activity occurred. It is generally related to an N- scope observations of the mineralogical and petro- NW deeping subduction of the Ionian oceanic litho- graphic characteristics. Selected thin sections were sphere which developed (probably since Middle-Late subsequently polished and carbon-coated for SEM- Eocene) beneath the Paleo-European-Iberian continen- EDX investigations. Also, small fragments of the tal margin which led, during the Oligocene, to the for- ashlars were attached to aluminium sample holders mation of the rift between Sardinia and Provence and carbon-coated. SEM-EDX instrumentation was a (Burrus 1984; Beccaluva et al. 1989; Beccaluva, Bian- Philips XL30 with Genesis EDS. chini, Coltorti, et al. 2005; Cherchi et al. 2008; Beccaluva Whole-rock chemistry of all samples was deter- et al. 2011). mined on selected portions (avoiding the larger miaro- The pyroclastic rocks used for the old Castro Cathe- litic cavities) by XRF with a Philips PW1400 dral belong to the first phase of Cainozoic Sardinian spectrometer using an Rh-tube for the analysis of volcanism (Beccaluva et al. 1989; Beccaluva, Bianchini, both major elements and some trace elements (Rb, Coltorti, et al. 2005; Beccaluva, Bianchini, Bonadiman, Sr, Pb, Zn, Y, Nb, and Zr), and a W-tube for the analysis et al. 2005; Beccaluva et al. 2011; Lustrino, Duggen, of the elements Ni, Cr, Ba, V, La, and Ce. The analyses and Rosenberg 2011), characterised by orogenic mag- were carried out on pressed-rock powders as matic activity developed mostly during Late Eocene– suggested by Franzini, Leoni, and Saitta (1972). The Miocene times (∼ 38–15 Ma). The major and trace measurement accuracy is±1% for SiO2, TiO2,Al2O3, element indicators and Sr–Nd–Pb–Hf–Os–O isotopes Fe2O3, CaO, K2O, and MnO and ±4% for MgO, Na2O, highlight complex petrogenetic processes including and P2O5; the accuracy of trace elements is ±2–3% to subduction-related metasomatism, hybridisation of 1000 ppm; ±5–10% at 100 ppm, and ±10–20% to the mantle and crustal melts, variable degrees of 10 ppm (Wangler and Scherer 2008; Columbu 2017). crustal contamination, and fractional crystallization at The amount of volatile components was determined shallow depths (Lustrino et al. 2013). by calculating the loss in weight % at 1000°C on The Sardinian orogenic magmatism starts in the powders dried at 105 ± 5°C until constant weight is north of the island in Calabona area (N Sardinia, ∼38 reached (L.O.I., Loss On Ignition). The FeO was deter- Ma (Lustrino et al. 2009)) with a small microdiorite mined by volumetric titolation with KMnO4 10N in outcrop, and it follows in Alghero (32.3 ± 1.5 Ma (Mon- acid solution. tigny, Edel, and Thuizat 1981)); (27.6 ± 1.5 Ma (Giraud, Whole rock powders were analysed by X-ray diffrac- Bellon, and Turco 1979)) and Osilo (31.2 ± 1.1 Ma (Mon- tion (XRD) with a PAN-ANALYTICAL diffractometer tigny, Edel, and Thuizat 1981)), and contempora- (anticathode Cu, tube voltage of 40 kV and current neously in the south of island in the Cixerri area (30.2 intensity of 30 mA) with an X-Celerator proportional ± 0.9 Ma (Beccaluva et al. 1985); 28.3 ± 1 Ma (Savelli counter. The identification of clay minerals was deter- 1975)) with andesitic lava, dacite domes, and rare mined on oriented and Mg2+-K+ saturated aggregates hypoabyssal bodies. of the <2 μm fraction. Mg2+ saturated specimens Starting from 22 Ma (Beccaluva et al. 1985; Speranza were scanned both in the air-dried and glycolated et al. 2002; Gattaceca et al. 2007; Carminati and states. For some selected Mg2+ saturated specimens, Doglioni 2012), a highly explosive fissural activity pro- glycerol treatment was also performed. K+ saturated duced abundant pyroclastic dacitic-rhyolitic products, mounts were run at room temperature as well as alternated with emissions of basaltic and andesitic after heating at 100, 300, and 550°C. lavas. The volcanic emissions occurred in various areas of Sardinia, prevalently along the western graben trending N-S (Figure 1). Geological setting of the Chilivani- The volcanic rocks of Oschiri Basin, which were used Berchidda Basin for Church of Nostra Signora di Castro, can be attributed to this later volcanic activity. Radiometric ages available Cenozoic Sardinian volcanism for volcanics in the Chilivani-Berchidda Basin area indi- The Oschiri area (Figure 1), where the Church of Nostra cate a 23–18 Ma range (Savelli et al. 1979; Oggiano, Signora di Castro is located, belongs to the Chilivani- Pasci, and Funedda 1995;Leccaetal.1997). The MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 401 pyroclastic materials erupted in the Oschiri Basin from subaerial volcanic centres. They were deposited in a fluvial-lacustrine environment (Vardabasso and Atzeni 1962; Cherchi and Montadert 1982; Sau et al. 2005). In this Oschiri area, except for the Mesozoic, other geological formations ranging from the Palaeozoic to the Cainozoic also outcrop (Figure 3).

Geographical zonation of the Chilivani-Oschiri area Within the western area of the Chilivani-Berchidda Basin, different facies of pyroclastic rocks are observed. On the base of macroscopic petro-volcanological features observed in the field at an early phase of research, this area was divided into two different sectors, an eastern one and a western one in respect to Lake Coghinas and the Rio Mannu river (Figure 3): the Oschiri sector, in which the Church of Nostra Signora di Castro is located (and other churches of Nostra Signora di Otti and San Demetrio), and the Chilivani- Tula sector, where many variable pyroclastic products outcrop. Chemical and petrographic analysis proved this field division to be accurate. Moreover, this division closely overlaps the subdivision between the old dioceses of Nostra Signora di Castro (Figure 4(a, b)), further east, and Sant’Antioco of Bisarcio (Figures 3 and 4(e, f)), further west. Volcanic rocks of the Oschiri and Chilivani/Tula Figure 4. Photographs of ignimbritic ashlars used for the sectors appear in vast outcrops, often made up of Castro Cathedral: (a) upper part of main west facade; (b) several superimposed cooling units with a maximum upper part of south facade; (c) main ignimbritic facies used total thickness of up to several hundred metres. The in the Church showing different colours of rock glassy matrix; Oschiri pyroclastic rocks have a clearly lower thickness (d) detail of column base in main facade, showing the differen- in respect to the more western Chilivani/Tula ones. tiated decay of welded and poorly welded volcanics, with exfo- fl Minor outcrops are often made up of one or more liation and aking processes, also as a function of the orientation of volcanic flow planes of rock (parallel and orthog- cooling units. onal to the facade); (e, f) alveolation decay process in different Field surveys with extensive sampling in the area of ignimbritic facies in the main facade, with evident chromatic Oschiri and Chilivani, together with the varying geo- alteration between the original surface of stone with more chemical and physical characteristics of the volcanic saturated colour (upper area of photo) and the actual surface fi rocks analysed (Columbu et al. 2014), highlight the of altered substrate (alveolated zone); (g) ssuring and fractur- ing processes with loss of material in the apse (east facade), presence of pyroclastic rocks with different petrophysi- where it is possible to observe the presence of ancient cal features. According to various authors (Smith 1960; plaster/treatment residues on the masonry, and the cement- Ragan and Sheridan 1972; Riehle 1973; Peterson 1979; based joint mortars (particularly harmful for preservation pur- Streck and Grunder 1995; Sparks, Tait, and Yanev 1999; poses) used in the last decades in restoration interventions; Quane and Russell 2005a; Pioli and and Rosi 2005; and (h) greenish poorly-welded ignimbritic facies with glassy Quane and Russell 2005b), these latter depend on matrix decohesion, where it is possible to observe the enu- cleated phenocrystals of sialic minerals and rare mafic crystals several volcanological factors: emplacement modes with very small size. and temperatures, cooling rates, welding and compaction degree, rates of loss of volatiles during and/or immediately after emplacement, alteration processes; Floyd 1977; Ghiara et al. 1997; Morbidelli et al. 1999; varying presence of ash, pumice (variably flattened Cerri and Oggiano 2002). and elongated), lithics (from sharp-cornered to Within the Oschiri sector, according to Columbu rounded), and crystals. (2017), three groups of rocks can be distinguished These pyroclastic rocks, being deposited in a (see Figure 3): less fractionated pyroclastics of the fluvial-lacustrine environment, were affected by western zone (LFW), intermediately-fractionated pyro- local processes of halmyrolysis and syn-epigenetic clastics of the central zone (IFC), and more fractionated transformations of the deposits (Winchester and pyroclastics of the eastern zone (MFE). The zones where 402 S. COLUMBU ET AL. the LFW and MFE volcanics outcrop are geographically and Floyd (1977), De La Roche et al. (1980), and Le separated by a central zone of IFC volcanics located Maitre et al. (2002). The rock classification is based on near the Oschiri station (Figure 3). However, field evi- bulk rock analysis of ignimbrite samples and it is dence seems to indicate that IFC volcanics were par- intended for a comparison with the chemical compo- tially covered by LFW and MFE volcanics. sition of the rocks used as building stones and not In the Chilivani/Tula sector, according to Columbu for petrogenetic purposes. (2017), there are two main pyroclastic facies: welded All the samples show porphyritic texture, with a por- pyroclastic rocks (WPC), similar to lava-like ignimbrites, phyritic index generally less than 20% (frequently where the pumices are small and rarely found, and between 9 and 15, Table 1), where the phenocrysts welded lapilli are always present; unwelded pyroclastic are moderately aligned in the direction of the rocks (UPC), cineritic facies of typical fall deposits, with flattened lapilli. The colour index varies from 5 to 7. Tex- a great compositional heterogeneity due to the vari- tures vary from vitroclastic to eutaxitic to parataxitic to able incidence of lithic fragments, pumice, and pheno- massive vitrophyric (Figure 5). crystals (e.g. large biotite). The mineral paragenesis mainly consists of opaque (i.e. magnetite and Ti-magnetite), plagioclase, and augitic clinopyroxene (often with a border of opaque Results granules) ± apatite ± sanidine ± quartz ± biotite. Xeno- The stones of the Nostra Signora di Castro crystals are also present, mainly represented by plagio- Church clase and pyroxene. The groundmass varies from hypohyaline to hyaline, affected by devitrification. Macroscopic features Devitrification structures (i.e. spherulites made up of The volcanic stones used to build the Church of Nostra radially-arranged acicular crystals), according to Signora di Castro (Figure 4) are pumiceous lapilli and Lofgren (Le Maitre et al. 2002) are sometimes present. ff ash pyroclastic rocks (ignimbrite) with di erent Ti-magnetite and (more rarely) hematite are present colours (from greenish to pinkish to orange-reddish) in the groundmass. The lithics consist of accidental and with a medium-high welding and consequent and/or cognate fragments (with andesitic or andesi- good physical-mechanical strength. The juvenile com- tic-dacitic composition). ponents of the rock are represented by lapilli, ash, Pumices are generally elongated and flattened. The fi and crystals. Lithic clasts consist of ne-grained lava iso-orientation of lithic clasts and crystals, and both the and ignimbrite fragments, with occasionally coarse- strong alignment of fiammae and the flattening of grained intrusive and subvolcanic rock fragments. The pumiceous components indicate a high compaction fl fi lapilli and the coarse ash are ame-like ( ammae) of the pyroclastic flow, involving the formation of fl and vary from moderately attened to elongate. On sub-parallel planar discontinuities inside the rock and fi the fresh cut of the rocks, ma c and sialic phases (i.e. variable glassy-matrix sintering. pyroxene and more rarely quartz) are visible with size In the most welded pyroclastic samples, the ground- less than a millimetre. Pumices (usually from greenish mass is compact, with matrix shards nearly adhering to ochre colours) are present. Small reddish-brown among them (where the intergranular porosity has a grains (<1 mm) are visible, probably belonging to small pore radius). The vesicles are collapsed and the fi altered ma c minerals. pumiceous clasts are strongly deformed. On the other On the outer surface of ashlar stones, the pumiceous hand, in medium welded samples, the sintering is lapilli are less visible with respect to the inside of the only partial with matrix shards moderately deformed, rock, due to their erodibility and subsequent physical- and a part of the pumices can still be recognised by mechanical removal. The small reddish-brown phases, their still round vesicles. also, are less visible on the outer surface. The ignimbrite According to the petrographic-textural features ashlars show the frequent presence of more or less observed in the microscopic analysis (e.g. welding – elongated vacuoles (frequent size: 2 8 mm) and degree, textural/structural parameters, porosity, etc.) – pores from irregular to sub-rounded (size: 1 5 mm), the following different pyroclastic facies used were possibly due to syngenetic degassing processes and identified: secondary alteration (probably at the expense of the pumice), respectively. 1) strongly welded pyroclastics (i.e. ignimbrites), with Petrographic features flattened and flame-like pumice (fiammae), The petrographic observations are summarised in flattened filled lithophysae and collapsed vesicles, Table 1, where there are also reported the Thorton where intergranular space is strongly reduced, and Tuttle differentiation index range (Thornton and referable to the IV-V welding ranks on the base of Tuttle 1960) and the rock classification on the chemical petrographic features (according to Quane and base according to the main diagrams of Winchester Russel classification (Quane and Russell 2005b)); Table 1. Summary scheme of the main petrographic and geochemical characteristics of the pyroclastites of Nostra Signora di Castro Cathedral and volcanic outcrops from Oschiri and Chilivani/Tula sectors. Geochemical characteristics Petrographic features by microscopic, SEM, and XRD analysis fi Lithotype group / Rock classi cation D.I. Porphyr. Origin of geographic Le Maitre (De La Roche (Winchester and (Thornton and index (vol Secondary samples zonation et al. 2002 et al. 1980) Floyd 1977) Tuttle 1960) Structurea %) Phenocrystsb Ground mass Texture mineralsc Ashlars from A1-A2-D TrDa, Da, Rhy, RhyDa, RhyDa, Da, An, 70–81 Porphyric 10–15 Opq, Pl, ± Opx, Cpx ± hyaline, hypohyaline with from vitroclastic to Fe-Phy, K-Fsp, Otti church (TrAn) QzLt, (TrAn), (Tr) Ap, ± Bt, rare Sa, ± crypto-crystalline eutaxitic to Cel, Glt B1-B2 (QzTr), (Tr) Qz assemblage of quartz and massive vitrophyric Fe-Phy, K-Fsp ±

K-feldspar Cal ALTERATI MINERALOGICAL-CHEMICAL C Fe-Phy, K-Fsp ± Cel ± Glt Outcrops from LFW TrDa, (Da), Rhy, RhyDa, RhyDa, Da, 70–78 Porphyric, 9–13 Opq, Pl, ± Opx, ± Cpx, hyaline, hypohyaline from vitroclastic to Fe-Phy, K-Fsp, Oschiri sector (Rhy) QzLt (TrAn), (An) granophyric ± Ap, ± Bt, ± Sa, ± eutaxitic to Cel, Glt Qz massive vitrophyric IFC Rhy, RhyDa, 76–79 10–14 Opq, Pl, ± Opx, ± Cpx, (QzLt) ± Ap, ± Bt, ± Sa, ± Qz MFE Rhy 77–82 11–15 Opq, Pl, ± Opx, ± Cpx, ± Ap, ± Bt, rare Sa, ±Qz Outcrops from WPC Rhy, Da, QzLt RhyDa, Da, 82–88 Porphyric 5–15 Opq, Pl, ± Opx, rare hyaline, hypohyaline from eutaxitic to Mor, silica phases Chilivani/Tula TrDa (TrAn) Cpx, KFsp, Qz rare vitrophyric (e.g. Opl-CT), sector Hbl Hm, Chl ONANDORIGINOFIGNIMBRITICSTONES UPC Rhy, RhyDa, 71–89 Porphyric 7–18 Opq, Pl, ± Opx, Cpx, ± hyaline, hypohyaline (QzLt) Bt, KFsp, Qz Mineral abbreviations: Qz = Quartz; Fsp = Feldspar; Opq = opaque minerals; Pl = Plagioclase; Opx = Orthopyroxene; Cpx = Clinopyroxene; Sa = Sanidine; Bt = Biotite; Hbl = Horneblende, Cel = Celadonite; Glt = Glauconite; Hm = Hematite; Cal = Calcite; Fe-Phy = Phyllosilicate; Chl = chlorite; Mor = Mordenite; Opl = Opal. Rock classiﬁcation abbreviations: Rhy = Rhyolite; RhyDa = Rhyodacite; Da = Dacite; QzLt = Quartz-Latite; QzTr = Quartz-Trachyte; Tr = Trachyte; TrAn = Trachy-andesite; An = Andesite. aAccording to increasing presence. bAccording to their increasing abundance. cAccording to their increasing abundance. 403 404 S. COLUMBU ET AL.

Church of Nostra Signora di Castro are classiﬁed as tra- chydacite and trachyte, and subordinately as trachyandesite with a sample into the dacite ﬁeld. The samples overlap the discriminant line of Irvine and Baragar (Winchester and Floyd 1977) distinguishing the alkaline from the subalkaline series (Figure 7(a)).

For a comparison, the R1 vs R2 diagram of De La Roche et al. (1980) was also used, because it is generally more in accordance with the petrographic classiﬁcation on a microscopic basis and because it includes more

chemical elements (R1 = 4Si-11(Na + K)-2(Fe + Ti); R2 = 6Ca + 2Mg + Al) with respect to the TAS diagram.

According to the R1–R2 diagram (modified; Figure 6) the ignimbritic samples of the Church of Nostra Signora di Castro are mainly classified as quartz-latite, rhyodacite, dacite and rhyolite and, subordinately, as quartz-trachyte, trachyte, and trachy-andesite. Moreover, considering the variable alteration degree of the rocks, the classification binary diagrams of Winchester and Floyd (1977), using some incompatible elements (Nb, Y, Zr), immobile during the alteration, were also used. According to these diagrams (Nb/Y vs

SiO2 wt%, Figure 8(a); Zr/TiO2 ppm vs SiO2 wt%, Figure 8(c); and Nb/Y ppm vs Zr/TiO2 ppm, Figure 8 (e)), the samples of the Church of Nostra Signora di Castro fall mainly in the rhyodacite, dacite, and andesite ﬁelds (Figure 8(a, c, e)) and, subordinately, in the trachyandesite ﬁeld (Figure 8c), highlighting their geo-

chemical features closer to those shown in the R1 vs R2 diagram (Figure 6). Figure 5. Microphotographs of thin sections (crossed nicols on For the reasons mentioned above, the diagram of the right, parallel nicols on the left) of ignimbrite rock samples De La Roche et al. (1980) was used as the main classifi- from IFC (a, b, c, d), LFW (e, f), and MFE (g, h), showing cation scheme of the analysed volcanic rocks. examples of the pyroclastic texture. (i, l): a lithic clast in the MFE ignimbrite rock. Scale: the long side of the photos is 5.1 mm except for C, D which are 1.35 mm. Field pyroclastic rocks Petrographic features 2) welded pyroclastics, with rare (or absent) fine- The juvenile clasts of the pyroclastites of the different grained phyllosilicates inside the porosity, referable subzones of Oschiri sector show relatively similar to the IV welding rank based on petrographic microscopic characteristics as regards their paragen- characteristics; and esis, porphyritic index (P.I. ∼ 9–15), and colour index 3) from slightly welded to unwelded pyroclastics, with (C.I. ∼ 2–6), Table 1. the presence of open lithophysae and fine-grained The phenocryst assemblage consists of plagioclase phyllosilicates lining porosity, referable to the (with labradorite to andesine compositions), ± clinopyr- III–IV welding ranks based on petrographic features. oxene (with augitic composition), ± sanidine ± quartz ± biotite, Fe-Ti oxides ± apatite as accessories. Xenocrys- Geochemical characteristics and rock tals with varying composition, principally plagioclase classification and pyroxene, are also present. Tables 2–4 report the chemical analyses of the samples The groundmass is from hyaline to hypohyaline. from the Church of Nostra Signora di Castro, where the Textures vary from vitroclastic to eutaxitic, to parataxi- alkali sum, the differentiation index (D.I. = normative Q tic, to massive vitrophyric. Locally, there are granophy- + Ab + Or + Ne + Kp + Lc; Thornton and Tuttle (1960) ric intergrowths of quartz and feldspars, indicating low and the alumina saturation index (A.S.I., defined as cooling speeds. the molecular ratio [Al2O3]/([CaO]–1.67 [P2O5]+ Two generations of plagioclase phenocrysts can be [Na2O] + [K2O]) are reported. recognised. The plagioclase crystals of the first gener- According to the Total Alkali Silica (TAS) diagram of ation (from subhedral to anhedral) show evident dise- Le Maitre et al. (2002)(Figure 7(a)), the samples of the quilibrium with the liquid, and are clearly zoned, Table 2. Chemical analysis of selected volcanics from the Church of Nostra Signora di Castro Cathedral. 1 2 3 4 5 6 7 8 9 10111213141516171819202122232425 Sample CO81S CO81A CO82S CO82A CO84S CO84A CO85S CO85A CO86S CO86A CO88A CO90S CO90A1 CO90A2 CO95S CO95A CO96S CO96A CO97S CO97A1 CO97A2 CO115S CO115A1 CO115A2 CO120A − − − − − − − − − − Subsig. A A+ A A+ A A+ A A+ A A+ A+ A A+ A+ A A+ A A+ A A+ A+ A A+ A+ A+ Class. R1-R2 qzLt qzLt qzLt RyDc Tr qzLt RyDc Dc qzLt qzLt qzLt qzTr Dc qzTr qzTr Lt qzLt qzLt qzLt Dc qzLt qzLt Dc RyDc qzLt SiO2 60.93 60.75 65.13 64.82 61.80 61.24 64.51 61.97 62.76 60.64 62.77 63.30 63.29 62.57 60.39 57.39 61.67 61.99 64.77 63.01 63.96 61.94 62.99 64.51 61.13 TiO2 0.79 0.79 0.64 0.66 0.64 0.67 0.66 0.73 0.66 0.64 0.66 0.71 0.68 0.65 0.71 0.68 0.84 0.77 0.70 0.70 0.71 0.65 0.67 0.68 0.81 Al2O3 16.10 16.43 14.06 13.88 13.70 13.70 14.01 13.63 14.01 13.53 13.81 14.47 14.47 14.10 14.68 13.80 15.92 15.55 15.23 13.66 14.59 12.43 13.41 13.21 15.93 Fe2O3 4.84 5.21 4.22 4.27 5.05 4.69 5.93 5.27 5.21 5.01 5.02 5.16 5.08 4.83 5.66 5.09 5.84 5.70 4.05 3.89 4.15 6.47 5.28 5.95 5.78 FeO 0.55 0.21 0.56 0.46 0.21 0.10 0.20 0.47 0.24 0.18 0.20 0.10 0.20 0.24 0.12 0.11 0.44 0.22 0.32 1.51 0.49 0.11 0.12 0.41 8.00 MnO 0.10 0.10 7.00 6.00 0.13 0.14 8.00 0.10 0.10 8.00 0.14 0.11 0.15 0.12 0.10 0.14 0.13 0.12 9.00 9.00 0.10 0.10 8.00 0.13 0.12 MgO 1.72 1.45 1.20 1.42 1.65 1.53 1.46 1.85 1.02 0.87 1.47 1.18 1.47 1.53 0.90 0.81 1.10 1.07 1.52 2.26 1.72 1.84 1.56 1.81 0.96 CaO 3.57 3.56 2.97 2.87 2.84 3.98 3.30 4.59 3.71 5.01 3.69 2.65 3.45 2.81 3.94 6.23 3.55 3.63 3.07 3.49 3.26 4.28 3.83 3.11 4.03 Na2O 3.87 3.80 3.79 3.69 4.00 3.04 3.52 3.65 3.94 3.91 3.92 3.64 1.69 4.07 4.05 3.49 3.82 3.65 4.10 3.49 4.20 3.52 3.53 3.11 4.17 K2O 5.17 5.37 4.45 4.15 6.92 6.85 3.95 3.79 4.81 4.65 5.12 6.49 6.60 6.16 5.55 4.95 5.60 5.65 4.36 4.55 4.20 4.73 4.34 4.69 4.78 P2O5 0.47 0.28 0.42 0.34 0.34 0.94 0.30 0.99 0.45 0.37 0.39 0.29 0.74 0.25 0.74 1.78 0.30 0.35 0.23 0.39 0.27 0.24 0.33 0.31 0.60 L.O.I. 1.88 2.03 2.49 3.37 2.73 3.13 2.00 2.96 3.11 5.09 2.81 1.91 2.19 2.68 3.15 5.53 0.78 1.29 1.56 2.97 2.34 3.70 3.87 2.08 1.61 Total 99.99 99.98 100.00 99.99 1001.00 1001.00 99.92 100.00 1002.00 99.98 100.00 1001.00 1001.00 1001.00 99.99 100.00 99.99 99.99 100.00 1001.00 99.99 1001.00 1001.00 100.00 100.00 S.I. 15.99 13.65 12.71 15.33 13.13 13.40 16.35 19.91 10.44 9.23 13.99 10.43 15.06 13.01 8.57 8.76 10.46 10.32 15.23 21.94 17.00 18.24 16.54 18.83 9.69

Mg Val. 0.38 0.35 0.33 0.37 0.38 0.39 0.32 0.39 0.27 0.25 0.36 0.31 0.35 0.37 0.24 0.24 0.26 0.26 0.41 0.45 0.42 0.36 0.36 0.36 0.24 ALTERATI MINERALOGICAL-CHEMICAL A.I. 0.74 0.73 0.79 0.76 1.03 0.91 0.72 0.74 0.83 0.85 0.87 0.90 0.69 0.95 0.86 0.80 0.78 0.78 0.75 0.78 0.79 0.88 0.78 0.77 0.76 Q 7.61 7.09 17.34 18.30 4.36 8.79 17.93 14.77 12.24 9.94 10.85 9.43 17.82 7.19 6.76 8.36 7.51 8.93 14.52 13.99 13.48 11.81 15.27 17.55 8.27 C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Or 30.55 31.73 26.30 24.52 40.89 40.48 23.34 22.40 28.42 27.48 30.25 38.35 39.00 36.40 32.80 29.25 33.09 33.39 25.76 26.89 24.82 27.95 25.65 27.71 28.25 Ab 32.74 32.15 32.07 31.22 31.94 25.72 29.78 30.88 33.34 33.08 33.17 30.80 14.30 34.44 34.27 29.53 32.32 30.88 34.69 29.53 35.54 29.78 29.87 26.31 35.28 An 11.29 11.91 8.21 9.05 0.00 3.51 10.76 9.61 6.34 5.63 4.96 3.98 12.28 2.01 5.49 7.37 9.75 9.36 10.28 8.17 8.55 4.15 7.93 8.23 10.63 Ac 0.00 0.00 0.00 0.00 1.68 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Di 1.23 1.31 1.21 1.05 4.08 3.69 1.11 2.44 2.40 4.13 3.64 2.18 0.00 3.62 2.15 2.77 1.50 1.65 1.34 2.72 2.33 5.16 3.08 1.73 1.33 Ed 1.58 1.99 1.99 1.43 5.91 4.78 1.98 3.16 5.39 10.25 5.45 3.99 0.00 5.03 5.76 7.44 3.54 3.79 1.54 2.77 2.55 7.92 4.42 2.62 3.33 Sl 2.81 3.30 3.20 2.49 9.99 8.47 3.09 5.61 7.79 14.38 9.09 6.17 0.00 8.65 7.91 10.21 5.04 5.43 2.88 5.49 4.88 13.09 7.50 4.36 4.66 En 3.71 3.00 2.43 3.05 2.22 2.10 3.12 3.47 1.43 0.25 1.97 1.93 3.66 2.13 1.24 0.73 2.05 1.90 3.16 4.37 3.20 2.19 2.46 3.70 1.77 Fs 5.44 5.21 4.59 4.75 3.69 3.13 6.36 5.16 3.67 0.71 3.39 4.06 6.35 3.39 3.81 2.26 5.56 5.01 4.17 5.10 4.03 3.85 4.04 6.42 5.08 Hy 9.16 8.22 7.02 7.80 5.91 5.23 9.48 8.63 5.09 0.96 5.36 5.99 101.00 5.52 5.06 2.99 7.61 6.91 7.34 9.47 7.23 6.04 6.50 10.12 6.86

Mt 0.94 0.94 0.83 0.82 7.00 0.83 1.06 1.00 0.94 0.90 0.90 0.91 0.91 0.88 1.00 0.90 1.09 1.02 0.76 0.96 0.81 1.14 0.93 1.10 1.01 ONANDORIGINOFIGNIMBRITICSTONES Il 1.50 1.50 1.22 1.25 1.22 1.27 1.25 1.39 1.25 1.22 1.25 1.35 1.29 1.23 1.35 1.29 1.60 1.46 1.33 1.33 1.35 1.23 1.27 1.29 1.54 Ap 1.09 0.65 0.97 0.79 0.79 2.18 0.69 2.29 1.04 0.86 0.90 0.67 1.71 0.58 1.71 4.12 0.69 0.81 0.53 0.90 0.63 0.56 0.76 0.72 1.39 D.I. 70.91 70.97 75.70 74.05 77.19 74.99 71.06 68.05 74.00 70.50 74.27 78.58 71.12 78.03 73.82 67.14 72.92 73.20 74.97 70.40 73.84 69.54 70.78 71.57 71.80 SAL 82.20 82.89 83.91 83.10 77.19 78.49 81.82 77.66 80.33 76.13 79.24 82.55 83.44 804.00 79.31 74.51 82.67 82.56 85.25 78.57 82.39 73.69 78.71 79.81 82.43 FEM 15.49 14.61 13.24 13.15 19.65 17.98 15.58 18.92 16.12 18.32 17.51 15.09 13.94 16.87 17.03 19.52 16.03 15.64 12.84 18.15 14.90 22.05 16.97 17.59 15.46 V 51.6 55.1 37.3 39.6 45.9 46.6 62.5 58.9 47.7 44.2 50.8 49.1 49.0 47.5 48.4 45.9 67.1 54.4 41.9 48.9 41.5 59.1 59.5 60.5 48.5 Ni 5.6 8.3 5.8 5.2 7.9 5.2 5.1 5.8 5.0 3.2 5.9 4.9 7.2 6.5 6.2 5.7 5.6 6.5 6.9 4.5 4.4 6.5 4.1 6.8 4.6 Zn 105.0 99.5 97.0 101.0 87.0 91.7 81.5 103.0 89.1 77.7 76.6 218.0 143.0 91.9 85.0 89.1 85.2 74.8 86.1 94.7 96.0 74.3 78.6 91.2 71.0 Rb 125 126 188 148 230 190 139 133 150 150 161 193 190 216 141 129 144 146 108 170 110 242 190 253 122 Sr 239 245 212 209 182 172 234 258 279 222 216 248 198 179 256 262 226 231 233 211 238 219 213 211 254 Ba 643 701 552 553 531 487 558 513 871 534 523 1010 538 535 590 519 564 558 610 582 621 486 543 509 564 Zr 213 226 195 194 193 182 190 176 186 181 180 187 189 192 190 176 181 187 220 201 212 174 179 178 183 Pb 20.0 35.6 n.d. 33.4 17.3 12.5 22.2 13.5 36.7 23.0 14.9 37.1 38.6 18.4 n.d. 19.6 15.1 28.8 27.3 21.4 n.d. 21.3 n.d. 17.0 n.d. Nb 12.5 9.5 9.1 12.8 9.5 8.6 9.5 12.1 6.8 12.6 9.2 11.1 12.1 11.6 8.4 7.8 12.3 10.5 10.9 9.0 12.0 9.5 13.8 11.1 10.8 Y 37.4 50.2 39.9 49.8 38.2 45.0 35.8 39.3 36.9 29.0 34.0 37.2 42.2 39.7 33.4 26.0 33.6 31.8 41.5 45.0 54.3 42.2 35.9 47.8 36.3 La 35.1 38.2 33.8 33.3 26.0 27.6 39.5 31.3 24.9 28.4 20.9 30.8 31.2 20.2 23.1 23.7 32.3 23.1 34.2 40.8 38.2 33.6 29.4 25.0 10.5 Ce 54.8 76.0 53.4 69.8 49.3 56.2 49.3 69.4 55.8 45.7 46.3 48.1 53.4 56.3 48.4 58.9 32.9 49.4 59.7 62.0 58.0 45.9 58.2 59.1 50.0

Major element data expressed in wt%, trace element data expressed in ppm. CIPW norm calculated setting Fe2O3/FeO wt% = 0.15. Abbreviations: Class. = classiﬁcation according to De La Roche et al. (1980); Rhy = Rhyolite; RhyDa = Rhyodacite; A.I. = agpaitic 405 index (molar [Al2O3]/([Na2O] + [K2O]) (Shand 1951); D.I. = normative Quartz + Albite + Orthoclase + Nepheline + Kaliophilite + Leucite, diﬀerentiation index of Thornton and Tuttle (1961); n.d. = not detected. 406

Table 3. Chemical analysis of selected volcanics from the Church of Nostra Signora di Castro Cathedral. 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Sample CO124S CO124A CO127S CO127A CO129A CO130S CO130A CO131S CO131A CO139A CO140A CO141S CO141A CO144S1 CO144S2 CO145S1 CO145A1 CO145S2 CO145A2 CO146A1 CO146A2 CO146A3 CO147A1 CO147A2 CO148S AL. ET COLUMBU S. − − − − − − − − − − Subsig. A A+ A A+ A+ A A+ A A+ A+ A+ A A+ A A A A+ A A+ A+ A+ A+ A+ A+ A Class. Dc Dc qzTr qzTr Dc Ry Dc Dc Dc Dc Dc RyDc RyDc qzTr qzLt TrAn TrAn qzLt Dc qzTr qzTr Dc Tr qzLt Tr

SiO2 64.11 63.40 61.99 61.96 63.75 66.77 61.24 63.17 62.82 62.56 64.23 65.30 64.55 60.11 60.16 57.74 56.96 59.81 61.29 60.60 601.00 61.96 57.24 60.93 59.76 TiO2 0.73 0.72 0.70 0.70 0.72 0.55 0.66 0.69 0.66 0.73 0.70 0.76 0.77 0.75 0.80 0.81 0.69 0.73 0.67 0.79 0.75 0.71 0.84 0.76 0.77 Al2O3 14.46 14.16 15.41 15.83 14.17 13.61 12.40 13.49 12.97 13.86 13.42 14.11 14.54 14.93 15.49 13.88 13.78 14.15 13.51 14.60 13.76 13.63 14.98 14.96 15.20 Fe2O3 5.29 5.51 5.56 5.28 4.98 4.55 6.19 5.51 6.13 6.11 5.96 5.68 3.90 5.93 5.82 5.75 5.46 5.68 4.92 6.03 6.03 5.45 6.81 5.75 6.00 FeO 0.50 0.86 0.10 0.23 0.90 0.10 0.12 9.00 0.23 0.15 0.30 0.38 1.40 8.00 0.29 0.46 0.21 0.46 0.56 0.11 9.00 0.11 0.11 7.00 0.24 MnO 0.13 0.13 0.12 0.11 0.13 5.00 0.10 8.00 9.00 9.00 0.10 7.00 0.11 7.00 6.00 0.12 0.17 0.11 0.13 9.00 9.00 9.00 0.15 0.10 0.12 MgO 1.27 1.43 1.05 0.88 1.16 0.90 1.39 1.44 1.56 1.32 1.72 1.38 1.22 0.76 0.69 1.31 1.16 1.26 1.00 1.47 1.42 1.16 1.21 0.70 1.55 CaO 3.04 3.12 3.23 3.50 2.97 2.17 4.94 4.19 3.84 4.48 3.46 3.12 3.14 3.91 4.24 4.93 5.01 3.63 4.10 3.67 4.35 4.46 4.45 4.23 4.00 Na2O 3.54 3.34 4.25 4.25 3.61 3.85 3.31 3.72 3.03 3.44 3.30 3.43 3.42 4.45 3.84 3.99 3.92 3.65 3.32 4.55 4.15 3.28 3.75 3.64 4.18 K2O 4.46 4.30 5.73 5.36 4.13 3.68 4.80 3.97 4.76 4.08 4.18 4.02 4.04 4.48 4.41 4.82 4.94 4.96 4.53 4.73 5.01 4.46 5.85 5.29 5.48 P2O5 0.24 0.24 0.21 0.23 0.32 0.17 0.25 0.23 0.21 0.35 0.30 0.24 0.25 1.27 1.45 1.03 1.01 1.51 1.15 0.39 0.41 0.58 1.15 0.77 0.27 L.O.I. 2.25 2.79 1.65 1.68 3.16 3.61 4.61 3.42 3.70 2.84 2.34 1.51 2.67 3.26 2.75 5.15 6.70 4.06 4.80 2.98 3.92 4.09 3.45 2.81 2.42 Total 1002.00 100.00 100.00 1001.00 100.00 1001.00 1001.00 100.00 100.00 1001.00 1001.00 100.00 1001.00 100.00 100.00 99.99 1001.00 1001.00 99.98 1001.00 99.99 99.98 99.99 1001.00 99.99 S.I. 13.70 15.77 9.52 8.39 13.03 10.68 14.63 15.77 16.68 14.93 18.70 15.63 14.06 7.84 7.72 12.94 11.58 12.77 11.30 13.67 13.42 13.03 11.19 7.27 13.83 Mg Val. 0.30 0.30 0.27 0.24 0.28 0.28 0.30 0.34 0.33 0.29 0.35 0.31 0.31 0.20 0.18 0.29 0.29 0.29 0.26 0.32 0.31 0.29 0.26 0.19 0.33 A.I. 0.74 0.72 0.86 0.81 0.73 0.76 0.86 0.77 0.78 0.73 0.74 0.71 0.69 0.82 0.72 0.85 0.86 0.80 0.77 0.86 0.89 0.75 0.83 0.78 0.84 Q 16.22 16.45 6.23 7.22 17.01 22.42 12.07 15.20 15.64 15.03 17.70 19.22 18.81 8.71 11.68 5.71 5.27 10.96 15.50 6.00 6.31 15.13 2.85 10.00 3.46 C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.16 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Or 26.35 25.41 33.86 31.67 24.40 21.75 28.36 23.46 28.13 24.11 24.70 23.75 23.87 26.47 26.06 28.48 29.19 29.31 26.77 27.95 29.60 26.35 34.57 31.26 32.38 Ab 29.95 28.26 35.96 35.96 30.54 32.58 28.01 31.48 25.64 29.11 27.92 29.02 28.94 37.65 32.49 33.76 33.17 30.88 28.09 38.50 35.11 27.75 31.73 30.80 35.37 An 10.39 10.94 6.05 8.29 10.26 8.99 4.80 8.39 7.73 10.33 9.46 11.23 12.39 7.53 11.56 5.73 5.41 7.58 8.58 5.44 4.12 9.30 6.76 8.86 6.53 Ac 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Di 0.91 0.86 2.26 1.80 0.64 0.18 5.23 3.53 3.08 2.74 1.91 0.81 0.47 0.72 0.00 3.43 3.55 0.16 1.09 3.18 4.49 2.58 1.96 1.39 3.64 Ed 1.74 1.65 5.09 4.69 1.39 0.39 10.12 5.68 5.35 5.43 2.94 1.46 0.85 2.35 0.00 6.77 7.39 0.32 2.54 5.46 8.03 5.11 4.71 4.74 6.13 Sl 2.65 2.52 7.34 6.49 2.03 0.57 15.35 9.20 8.42 8.17 4.84 2.27 1.32 3.08 0.00 10.20 10.93 0.48 3.63 8.64 12.53 7.69 6.67 6.13 9.77 En 2.74 3.16 1.57 1.36 2.59 2.16 1.04 1.95 2.46 2.02 3.40 3.06 2.82 1.56 1.72 1.67 1.25 3.06 1.99 2.19 1.45 1.69 2.10 1.10 2.17 Fs 6.04 6.95 4.06 4.07 6.43 5.31 2.30 3.61 4.90 4.59 6.01 6.38 5.89 5.81 7.13 3.78 2.98 7.23 5.31 4.31 2.98 3.84 5.80 4.32 4.20 Hy 8.79 10.11 5.63 5.42 9.02 7.46 3.33 5.56 7.36 6.61 9.41 9.44 8.71 7.37 8.85 5.46 4.22 10.30 7.30 6.50 4.43 5.54 7.90 5.42 6.37 Mt 1.01 1.11 0.98 0.95 1.03 0.80 1.09 0.97 1.10 1.08 1.09 1.05 0.94 1.04 1.06 1.08 0.98 1.07 0.96 1.06 1.06 0.96 1.20 1.00 1.08 Il 1.39 1.37 1.33 1.33 1.37 1.04 1.25 1.31 1.25 1.39 1.33 1.44 1.46 1.42 1.52 1.54 1.31 1.39 1.27 1.50 1.42 1.35 1.60 1.44 1.46 Ap 0.56 0.56 0.49 0.53 0.74 0.39 0.58 0.53 0.49 0.81 0.69 0.56 0.58 2.94 3.36 2.39 2.34 3.50 2.66 0.90 0.95 1.34 2.66 1.78 0.63 D.I. 72.53 70.12 76.05 74.85 71.96 76.74 68.44 70.14 69.41 68.24 70.32 72.00 71.62 72.84 70.23 67.95 67.63 71.15 70.36 72.45 71.03 69.24 69.15 72.06 71.21 SAL 82.92 81.07 82.09 83.13 82.22 85.73 73.25 78.52 77.14 78.57 79.78 83.23 84.01 80.37 81.96 73.68 73.04 78.73 78.94 77.90 75.15 78.54 75.92 80.91 77.73 FEM 14.39 15.67 15.77 14.73 14.19 10.27 21.61 17.57 18.63 18.06 17.37 14.77 13.01 15.85 14.79 20.66 19.79 16.73 15.81 18.60 20.39 16.88 203.00 15.78 19.31 V 52.5 53.8 66.2 63.6 67.2 33.1 53.9 51.0 62.2 58.0 53.9 52.3 53.1 41.7 44.2 61.3 54.8 94.1 69.6 54.1 43.7 40.9 67.0 49.4 59.9 Ni 5.1 6.3 5.4 4.4 5.4 4.5 5.2 4.5 4.5 4.9 5.7 5.7 7.2 6.6 5.1 3.3 6.1 3.7 6.9 3.6 6.2 4.3 6.3 4.3 6.9 Zn 82.7 90.7 61.9 68.1 79.3 93.1 93.1 74.4 73.8 80.1 85.1 79.8 76.0 62.6 64.2 102.0 106.0 77.7 82.6 133.0 109.0 99.9 94.7 65.1 86.8 Rb 139 145 152 141 158 131 221 134 248 127 194 143 131 128 126 178 176 169 153 160 169 166 151 144 159 Sr 228 211 249 232 226 240 200 233 218 223 223 228 230 277 303 261 255 219 215 249 267 249 243 239 259 Ba 572 567 560 585 552 615 495 526 489 550 549 546 560 603 639 436 496 544 477 569 565 566 509 591 552 Zr 190 191 203 200 176 194 194 183 173 188 181 190 194 201 204 178 178 194 176 196 195 188 161 192 197 Pb 14.2 35.5 n.d. n.d. n.d. n.d. 12.6 15.0 27.7 20.9 24.5 n.d. n.d. 19.2 n.d. n.d. 12.7 n.d. 12.8 26.4 n.d. 23.7 41.9 13.8 20.7 Nb 11.7 8.0 11.7 12.4 11.3 12.1 12.0 10.3 7.1 12.6 11.9 9.1 10.9 11.2 11.8 10.7 10.1 14.2 8.3 10.1 10.9 12.6 12.4 10.3 11.6 Y 39.5 41.6 42.5 37.1 36.4 39.6 41.5 40.8 35.8 48.5 31.5 35.5 40.3 43.8 33.6 29.9 33.9 46.8 46.4 32.7 35.6 43.9 31.0 42.3 47.9 La 30.2 47.8 34.5 30.1 24.9 45.6 34.5 29.8 26.5 30.3 24.3 34.4 25.1 26.8 33.7 34.3 18.2 29.8 38.6 25.7 23.1 33.5 19.7 29.0 32.9 Ce 43.0 59.5 59.4 56.8 51.2 38.4 47.2 60.7 47.7 48.8 49.1 44.0 64.9 61.8 55.0 160.0 45.7 52.9 63.3 56.8 38.5 50.7 46.5 53.3 59.0

Major element data expressed in wt%, trace element data expressed in ppm. CIPW norm calculated setting Fe2O3/FeO wt% = 0.15. Abbreviations as Table 2. Table 4. Chemical analysis of selected volcanics from the Church of Nostra Signora di Castro Cathedral. 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 Sample CO148A CO151A CO153S CO153A CO154A CO155S CO155A CO156S1 CO156S2 CO156A CO157A CO158S CO158A CO159S CO159A CO160S CO160A CO162S CO162A CO230S CO230A CO231S CO231A CO232S CO232A − − − − − − − − − − − Subsig. A+ A+ A A+ A+ A A+ A A A+ A+ A A+ A A+ A A+ A A+ A A+ A A+ A A+ Class. Lt TrAn qzLt qzLt Dc qzLt qzLt qzLt qzLt Dc Dc TrAn qzLt RyDc RyDc qzLt qzLt Dc Dc Tr qzLt qzLt qzLt qzLt Lt

SiO2 58.94 58.61 64.01 62.55 63.78 64.87 64.11 64.37 64.52 65.61 61.35 58.68 59.87 65.42 65.26 62.91 63.85 64.89 63.69 60.20 62.49 61.35 62.01 61.70 59.53 TiO2 0.70 0.70 0.72 0.69 0.69 0.71 0.72 0.64 0.66 0.70 0.72 0.77 0.74 0.70 0.71 0.66 0.71 0.75 0.74 0.70 0.50 0.73 0.68 0.84 0.78 Al2O3 14.85 14.42 14.02 13.65 14.39 15.03 14.47 13.66 13.77 14.72 14.11 14.70 14.87 13.71 13.91 14.23 14.11 14.81 14.34 13.54 13.71 14.31 14.45 16.00 15.24 Fe2O3 5.51 5.34 5.34 4.97 4.53 4.79 4.69 5.38 5.18 3.48 5.38 5.61 5.21 5.64 5.19 2.63 4.56 5.29 4.95 5.33 3.67 5.40 5.14 5.64 5.34 FeO 0.34 0.11 0.27 0.49 0.59 0.36 0.97 0.36 0.36 0.38 0.12 0.33 0.51 0.41 0.39 1.50 1.04 0.51 0.60 0.12 0.34 0.27 0.20 0.59 0.52 MnO 0.20 0.12 0.12 0.15 7.00 9.00 9.00 6.00 6.00 4.00 0.10 0.14 0.15 6.00 6.00 8.00 8.00 0.11 0.17 0.12 0.10 8.00 9.00 0.13 0.16 MgO 1.73 1.45 1.47 1.76 1.49 0.82 0.97 1.44 1.38 0.87 1.32 1.68 1.80 1.23 1.28 0.87 1.09 1.03 1.02 1.74 0.60 1.13 1.06 0.99 0.93 CaO 5.05 5.10 3.34 3.75 3.83 2.71 2.76 2.94 2.93 3.31 4.64 4.87 4.26 3.05 3.01 3.45 3.52 3.09 3.22 4.25 2.34 4.00 3.76 3.53 5.37 Na2O 3.66 4.54 3.35 3.35 3.60 3.70 3.53 4.08 4.10 4.12 3.81 4.52 4.47 3.36 3.59 3.85 3.79 3.65 3.61 3.41 3.81 3.72 3.85 3.72 3.35 K2O 5.38 5.09 5.43 5.36 4.02 5.14 5.12 4.04 4.00 4.13 4.02 4.65 4.52 4.48 4.42 4.10 4.60 4.21 4.20 6.47 5.38 4.47 4.51 5.52 5.34 P2O5 1.17 0.22 0.30 0.43 0.32 0.24 0.33 0.21 0.22 0.27 0.28 0.23 0.31 0.21 0.22 0.89 0.65 0.23 0.22 0.33 0.24 0.65 0.63 0.45 1.96 L.O.I. 2.47 4.30 1.62 2.84 2.69 1.55 2.24 2.82 2.82 2.37 4.14 3.83 3.28 1.72 1.95 4.84 2.01 1.42 3.25 3.78 6.82 3.89 3.62 0.89 1.48 Total 100.00 100.00 99.99 99.99 100.00 1001.00 100.00 100.00 100.00 100.00 99.99 1001.00 99.99 99.99 99.99 1001.00 1001.00 99.99 1001.00 99.99 100.00 100.00 100.00 100.00 100.00 S.I. 16.06 13.09 14.34 16.81 16.36 8.49 108.00 15.06 14.56 9.54 14.43 15.48 16.68 13.56 13.78 9.86 11.50 11.59 11.55 14.97 6.13 12.12 11.25 9.68 9.67 IEAOIA-HMCLALTERATI MINERALOGICAL-CHEMICAL Mg Val. 0.37 0.34 0.34 0.39 0.36 0.24 0.25 0.33 0.33 0.31 0.32 0.36 0.38 0.29 0.31 0.29 0.27 0.26 0.26 0.39 0.23 0.28 0.28 0.24 0.24 A.I. 0.80 0.90 0.81 0.83 0.71 0.78 0.78 0.81 0.80 0.76 0.75 0.85 0.82 0.76 0.77 0.76 0.79 0.71 0.73 0.93 0.88 0.77 0.78 0.76 0.74 Q 5.67 1.61 13.80 12.13 16.41 15.15 15.11 15.52 15.89 17.42 12.26 2.65 5.08 18.67 17.75 16.83 14.95 17.38 16.63 5.43 13.52 12.54 12.85 8.65 9.70 C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Or 31.79 308.00 32.09 31.67 23.75 30.37 30.25 23.87 23.64 24.40 23.75 27.48 26.71 26.47 26.12 24.23 27.18 24.88 24.82 38.23 31.79 26.41 26.65 32.62 31.55 Ab 30.97 38.41 28.34 28.34 30.46 31.31 29.87 34.52 34.69 34.86 32.24 38.24 37.82 28.43 30.38 32.58 32.07 30.88 30.54 28.85 32.24 31.48 32.58 31.48 28.34 An 8.20 3.93 7.18 6.38 11.23 9.22 8.52 7.03 7.36 9.47 9.53 6.09 7.16 9.10 8.79 9.44 7.90 11.59 10.52 2.53 4.42 9.15 8.83 10.66 10.78 Ac 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Di 3.15 6.56 2.42 3.41 1.98 0.64 0.75 1.95 1.82 1.63 3.63 5.63 4.28 1.30 1.43 0.53 1.43 0.57 1.04 5.97 1.26 1.79 1.57 0.93 0.72 Ed 4.60 10.26 3.85 4.49 2.80 1.64 1.84 3.28 3.04 2.76 6.24 8.33 5.75 2.67 2.56 1.05 3.10 1.34 2.39 7.78 3.57 3.70 3.29 2.43 1.90 Sl 7.75 16.82 6.27 7.90 4.77 2.27 2.58 5.23 4.86 4.38 9.87 13.96 103.00 3.97 3.99 1.58 4.53 1.91 3.43 13.75 4.83 5.49 4.86 3.36 2.62 En 2.85 0.57 2.54 2.80 2.79 1.75 2.07 2.68 2.59 1.41 1.60 1.57 2.50 2.46 2.53 1.92 2.05 2.30 2.06 1.57 0.91 1.98 1.91 2.04 1.98 Fs 4.76 1.03 4.64 4.24 4.54 5.17 5.83 5.16 4.98 2.75 3.16 2.67 3.84 5.81 5.21 4.33 5.09 6.20 5.45 2.34 2.96 4.71 4.58 6.11 6.03 Hy 7.61 1.60 7.18 7.04 7.34 6.92 7.90 7.84 7.57 4.16 4.77 4.25 6.34 8.27 7.74 6.25 7.14 8.51 7.51 3.91 3.87 6.70 6.49 8.15 8.01 Mt 1.02 0.94 0.97 0.95 0.89 0.89 0.99 1.00 0.96 0.67 0.95 1.03 1.00 1.05 0.97 0.74 0.99 1.01 0.97 0.94 0.70 0.98 0.92 1.09 1.02 ONANDORIGINOFIGNIMBRITICSTONES Il 1.33 1.33 1.37 1.31 1.31 1.35 1.37 1.22 1.25 1.33 1.37 1.46 1.41 1.33 1.35 1.25 1.35 1.42 1.41 1.33 0.95 1.39 1.29 1.60 1.48 Ap 2.71 0.51 0.69 1.00 0.74 0.56 0.76 0.49 0.51 0.63 0.65 0.53 0.72 0.49 0.51 2.06 1.51 0.53 0.51 0.76 0.56 1.51 1.46 1.04 4.54 D.I. 68.43 70.10 74.23 72.14 70.63 76.83 75.23 73.91 74.22 76.68 68.25 68.37 69.61 73.57 74.25 73.63 74.20 73.14 71.99 72.51 77.54 70.43 72.08 72.74 69.60 SAL 76.63 74.04 81.42 78.52 81.86 86.05 83.75 80.94 81.57 86.15 77.77 74.46 76.77 82.67 83.03 83.07 82.10 84.73 82.51 75.04 81.96 79.57 80.91 83.40 80.38 FEM 20.42 21.19 16.49 18.20 15.06 11.99 13.61 15.77 15.16 11.17 17.61 21.23 19.49 15.11 14.55 11.89 15.51 13.38 13.82 20.70 10.90 16.06 15.02 15.22 17.68 V 63.1 40.7 60.9 56.7 41.4 40.8 47.6 63.7 59.6 59.3 53.9 42.9 41.0 64.6 62.0 41.2 46.0 57.5 60.5 43.3 25.2 48.1 48.7 63.2 62.4 Ni 4.6 5.2 6.2 6.4 6.1 5.8 6.0 4.5 4.5 4.9 4.1 4.1 6.5 4.7 5.9 3.6 4.5 6.6 5.4 7.0 4.9 6.4 5.5 6.8 6.8 Zn 72.9 73.4 77.3 83.0 92.7 66.6 75.8 71.6 72.3 62.7 77.3 70.7 71.6 67.6 72.1 98.4 87.6 76.5 72.6 86.4 66.1 81.3 74.3 84.6 115.0 Rb 157 140 166 166 130 166 155 180 181 132 88.2 131 121 198 181 123 192 129 129 180 183 163 169 132 122 Sr 261 245 207 206 221 207 210 200 209 225 288 242 269 201 200 223 234 236 227 191 168 238 231 215 246 Ba 546 565 544 535 567 561 517 477 470 516 527 582 598 538 548 570 519 571 581 488 569 554 546 518 530 Zr 202 188 186 187 204 206 211 180 183 192 187 202 202 196 188 207 195 189 192 190 206 184 197 184 172 Pb 18.6 n.d. 25.0 21.9 14.3 n.d. 14.7 18.0 n.d. 23.0 18.8 15.6 n.d. n.d. n.d. 31.3 n.d. 15.3 12.0 25.9 28.7 n.d. 19.7 n.d. n.d. Nb 10.7 9.1 10.1 11.5 12.7 9.6 14.4 10.5 8.9 11.7 7.5 12.5 12.5 8.9 11.6 12.6 13.7 7.5 8.7 11.1 9.8 12.6 12.7 11.7 6.0 Y 43.8 44.8 42.4 38.2 37.4 46.4 47.4 39.1 41.5 38.6 38.1 42.2 43.3 45.8 39.9 31.8 43.2 34.7 44.2 41.6 48.7 34.1 49.8 35.9 24.2 La 30.5 26.8 31.3 25.3 18.6 32.1 36.8 24.7 29.9 28.7 31.4 30.5 17.9 25.8 35.0 26.5 32.2 28.4 38.6 43.2 36.2 20.4 39.3 21.7 22.6 Ce 64.6 52.1 58.2 52.8 54.5 60.1 65.6 51.0 46.3 50.3 41.0 45.9 43.2 52.8 53.6 53.9 56.7 58.4 76.4 48.9 60.9 52.2 68.3 53.1 52.9

Major element data expressed in wt%, trace element data expressed in ppm. CIPW norm calculated setting Fe2O3/FeO = 0.15. Abbreviations as in Table 2. 407 408 S. COLUMBU ET AL.

occasionally with oscillatory zoning, showing resorption and include (Ti–) magnetite microphenocrystals (<1 mm) and, rarely, clinopyroxene. They are often broken, with sharp edges. In the more evolved products, they are sometimes surrounded by neoformation rims in equilibrium with the liquid. In some cases, the plagioclase of the second generation appears weakly zoned with from euhedral to subhedral habitus, without any signs of disequilibrium. They sometimes include opaque oxides of dimensions <200 µm. As regards the size of plagioclases, all analysed samples show a bimodal size distribution with larger dimensions (<4 mm) of the first generation of plagioclase and smaller dimensions (<2 mm) of the later ones. Ti–magnetite occurs as microlites (≤0.2 mm), present in the groundmass and as microphenocrystals (<1 mm), included in the various phenocrystals or scattered in the groundmass. Besides these overall similar characteristics, some differences between the volcanics of the sub-zones of Oschiri sector are present. The less fractionated western pyroclastites (LFW) show on average a porphyritic index around 11 (Table 1; Figure 5(e, f)) and a colour index around 6. The paragenesis consists, in addition to plagioclases, of rare augitic clinopyroxene (often with a border of opaque granules), microphenocrystals of (Ti-) magnetite, ± apatite ± sanidine ± quartz ± biotite. Abundant Figure 6. Total Alkali Silica (T.A.S.) classification diagram of Le microgranules of Ti-magnetite ± hematite are present fi Maitre et al. (2002, modi ed): Na2O+K2O versus SiO2 (wt%). in the groundmass. There are lithic, accidental and/or fi (a) samples from the Castro Church; (b) samples from eld. cognate fragments (generally with rounded edges) Graphic symbols: crosses = monument samples from this work; squares = outcrop samples from Oschiri sector; rhombus with an andesitic or andesitic-dacitic composition = outcrop samples from Chilivani/Tula sector (WPC, UPC that contain plagioclase and pyroxene microlites. pyroclastics). Pumices are rare (than in MFE samples), generally

Figure 7. R1 vs. R2 classiﬁcation diagram of De La Roche et al. (1980, modiﬁed) with volcanic samples from the Oschiri sector and Santa Maria Otti Church. Graphic symbols: crosses = samples from Castro Church; empty triangles = samples from nearby Otti Church; black circles = outcrop MFE samples from Oschiri sector; rhombus = IFC samples from Oschiri sector; stairs = outcrop LFW samples from Oschiri sector; × = outcrop WPC samples from Chilivani/Tula sector; = outcrop UPC samples from Chilivani/ Tula sector. MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 409

Figure 8. Chemical classification of volcanic samples from the Nostra Signora di Castro Cathedral and outcrops from Oschiri and Chilivani/Tula sectors, according to the different diagrams of Winchester and Floyd (1977, modified). (a, b) Nb/Y (ppm) versus SiO2 (wt%); (c, d) Zr/TiO2 (ppm) versus SiO2 (wt%); (e, f) Nb/Y (ppm) versus Zr/TiO2 (ppm). Graphic symbols: crosses = Castro Church samples; squares = outcrop samples from Oschiri sector; rhombus = outcrop samples from Chilivani/Tula sector (WPC, UPC pyroclastics). elongated and flattened, and often include crystals Pumices are more abundant with respect to LFW having homogeneous composition. samples and are only slightly flattened and elongated. The intermediately-fractionated central pyroclastites The more-fractionated eastern pyroclastites (MFE) (IFC) show petrographic characteristics that lie show an average porphyritic index around to 13 and between the two other groups (Table 1; Figure 5(a– a colour index around 4 (Table 1; Figure 5(g, h)). The d)). They have a prophyritic index ranging between phenocryst assemblage is made up of plagioclases, 10 and 14, with a colour index of about 5. The pheno- those of the last generation being most common, cryst paragenesis consists of first- or second-generation microphenocrystals of Ti-magnetite ± clinopyroxene ± plagioclases in more or less equal ratios, micropheno- apatite ± sanidine ± quartz ± biotite. Lithic fragments, crystals of Ti-magnetite, rare clinopyroxenes, very rare rare or sometimes absent, have sharp or slightly sanidine ± apatite ± quartz ± biotite. In these pyroclas- rounded edges, and are sometimes of subvolcanic tic rocks, the lithic fragments are rare, show a basic- origin. Pumices are more abundant than in LFW intermediate composition and are slightly rounded. samples; they are flattened and more elongated. 410 S. COLUMBU ET AL.

The WPC rocks of Chilivani-Tula sector show variable they are mainly classified (according to De La Roche petrographic features (Table 1), even within the same et al. diagram (1980)) as rhyolite and rhyodacite and cooling unit. They have a porphyritic index between as subordinate quartz-latite (Figure 7); the MFE pyro- 5% and 15% for phenocrystals (in order of segre- clastics (eastern zone) have a D.I. of ∼77–82 (Tables 1 gation) of spinel, plagioclase, ± orthopyroxene, clino- and 5; Figure 9) and they are classified as rhyolites pyroxene, K-feldspar, quartz, and very rare (Figure 7). horneblende. There are angular or rounded cognate According to Columbu (2017), the WPC rocks of the fragments (2–7%). Nodules of gabbro and granoblas- Chilivani/Tula sector (including the lava-like ignim- tic rock fragments of the basement are occasionally brites), normally outcrop further to the NE with present. The UPC rocks show variable compositional respect to the UPC rocks, showing a D.I. range of characteristics due to the different presence of the ∼84–88 (Tables 1 and 6), are the most fractionated phenocrystals and pumice and lithic components. ones in the area and complete the trend displayed by The phenocrystals are plagioclases, ± orthopyroxene, products from the Oschiri sector (Figure 9). They are clinopyroxene, ± biotite, K-feldspar, and quartz. mainly classified as quartz-trachytes, and subordinately Biotite is often deformed and partially transformed as quartz-latites and rhyolites (Figure 7). The UPC rocks in chlorite minerals. have a D.I. range of ∼70–83 (Tables 1 and 6). Compar- ing the samples with same D.I., they show with respect

Geochemical characteristics and rock to WPC rocks (Figure 9) significantly lower TiO2,Al2O3, classification FeOt, MgO, Na2O, P2O5, Zn, Zr, Nb, and Y content, and The results of XRF chemical analysis of the volcanic out- higher SiO2, L.O.I. and Sr content (Table 6). They are crops from the Oschiri and Chilivani/Tula sectors are mainly classified as rhyodacites and subordinately as shown in the Tables 5 and 6, respectively. rhyolites (Figure 7). The whole-rock geochemical composition of an Observing the other classification diagrams, accord- ignimbrite deposit can be remarkably different from ing to Total Alkali Silica of Le Maitre et al. (2002), the magma composition and can vary across the deposit outcrop samples of Oschiri sector mainly show a tra- as a consequence of the eruption and emplacement chydacitic or dacitic composition, similarly to those of dynamics of these volcanic products. Nevertheless, the Church of Nostra Signora di Castro (Figure 6(b)). for the purposes of this research the major and trace The samples of Chilivani/Tula sector show two element chemical composition of the ignimbrite out- different populations: the samples classified as trachy- crops can be very useful for a comparison with the dacite falling into the alkaline series, and the samples rocks used in the Church of Nostra Signora di Castro, classified as dacite and rhyolite falling into the subalka- and with this finality the chemical analyses are line series (Figure 6(b)). reported in geochemical classification diagrams. According to the Wincester and Floyd diagram On the base of geochemical data, some differences (1977)(Figure 8(b)) the volcanics of Oschiri and Chili- emerged between the different volcanic facies and vani/Tula outcrops show a dacitic and rhyodacitic defined sectors. The WPC rocks of Chilivani/Tula and composition. the pyroclastites of Oschiri exhibit, on average, higher TiO ,AlO , FeO , and Na O and lower K O contents 2 2 3 tot 2 2 Mineralogical and chemical investigations on than the UPC as well as slightly lower Zr, Nb, Y, La, alteration and Ce concentrations. The magmatic evolution of volcanics and their geo- The study of the alteration processes of the Church’s chemical/geographic distributions in the Oschiri and volcanics has been dealt on the mineralogical basis,

Chilivani/Tula sectors are shown in the D.I. vs TiO2 through optical polarised microscopy (OM) of thin sec- diagram (Figure 9, D.I. = differentation index of Thorn- tions and XRD / SEM-EDX analysis, in order to investi- ton and Tuttle (1960). The WPC (welded pyroclastites gate the presence of accessory and / or secondary from the M. Sassu area) align with the trend of the phases not recognisable at the polarising microscope. most eastern volcanics (LFW, IFC, MFE) from the Then, on a chemical basis, a data comparison Oschiri sector. According to this latter diagram and between the outer sample (signed with A+) taken on the classification diagram of De La Roche et al. the external surface of the monument rock (presum- (1980), the volcanics from the two sectors show the fol- ably more altered) and the inner ‘fresh’ sample (A−) lowing characteristics. was made. As regards the Oschiri sector: the LFW pyroclastics (western zone) have a D.I. range of ∼70–78 (Tables 1 OM / SEM-EDX / XRD analyses and 5; Figure 9) and they are mainly classified as rhyo- By optical microscopy analysis, the groundmass glass lite and quartz-latite and as subordinate rhyodacite of the ignimbritic samples is often transformed in (Figure 7); the IFC pyroclastics (central zone) show a hardly recognisable cryptocrystalline minerals, reveal- D.I. range of ∼76–79 (Tables 1 and 5; Figure 9) and ing devitrification processes (Figure 5) with the Table 5. Chemical analysis of selected volcanics from Oschiri sector. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Sample CO3 CO3 CO5 CO5 CO6 CO18a CO18a O136b O136b CO197 CO210 CO211 CO212 CO222 CO1 CO1 CO2 CO2A CO2A CO8 CO14 CO199 CO199 CO221 CO221 − − − − − − − − − − − − − − − − − − Subsig. A A+ A A+ A A A+ A A A A A A A A A+ A A+ A+ A A A A+ A A Class. qzLt qzLt qzLt qzLt qzLt RyDc RyDc qzLt RyDc qzLt RyDc RyDc qzLt qzLt Ry Ry qzLt qzLt qzLt Ry RyDc RyDc RyDc Ry Ry SiO2 63.13 63.81 65.30 64.14 61.71 66.26 67.67 65.09 66.72 62.94 65.10 65.84 65.57 63.37 66.18 65.65 65.36 64.91 64.99 67.79 67.07 67.05 66.27 66.97 66.98 TiO2 0.76 0.73 0.72 0.64 0.84 0.68 0.68 0.74 0.67 0.73 0.74 0.73 0.74 0.71 0.70 0.66 0.70 0.65 0.70 0.68 0.68 0.69 0.71 0.70 0.69 Al2O3 14.95 14.75 14.72 12.88 14.77 14.82 14.38 15.16 14.04 14.34 14.32 14.41 14.55 15.61 14.22 14.09 14.58 14.43 14.67 14.27 14.41 14.95 14.69 14.51 14.47 Fe2O3 5.32 5.05 4.93 6.66 5.64 4.94 4.16 5.06 4.24 5.41 5.56 5.65 5.20 5.28 4.79 4.92 5.12 4.73 4.68 4.62 4.95 4.34 4.32 4.95 4.64 FeO 0.16 0.28 0.54 0.97 0.48 0.46 0.47 0.66 1.01 0.23 0.42 0.43 0.30 0.24 0.61 0.65 0.25 0.39 0.35 0.26 0.44 0.23 0.59 0.21 0.42 MnO 0.12 0.13 0.10 0.14 0.12 8.00 7.00 0.12 0.12 0.10 6.00 6.00 8.00 0.14 0.13 0.15 0.12 0.11 0.12 6.00 9.00 0.11 0.13 8.00 9.00 MgO 0.99 0.89 1.11 1.75 1.30 0.85 0.58 0.66 1.01 1.33 1.16 0.85 1.02 0.85 0.78 0.83 0.79 0.95 0.76 0.63 0.75 0.68 0.79 0.68 0.68 CaO 3.31 3.10 2.78 2.35 3.25 2.62 2.74 2.93 2.81 4.67 2.86 2.85 2.89 2.91 2.76 2.68 2.68 2.70 2.85 2.53 2.66 2.48 2.62 2.59 2.58 Na2O 3.21 3.02 3.81 2.86 2.92 3.71 4.13 3.80 3.69 4.86 3.26 3.57 3.75 3.46 3.73 3.84 3.18 3.19 3.18 4.04 3.97 3.84 3.79 3.76 3.73 K2O 6.35 6.42 4.58 5.53 5.27 4.02 4.02 4.36 3.66 4.06 4.48 4.33 4.48 4.72 4.67 4.28 6.23 5.92 5.95 3.88 3.77 3.86 3.89 4.40 4.31 P2O5 0.58 0.46 0.21 0.19 0.26 0.15 0.16 0.22 0.18 0.27 0.62 0.27 0.30 0.23 0.24 0.28 0.28 0.31 0.44 0.17 0.18 0.19 0.20 0.38 0.30 L.O.I. 1.12 1.37 1.20 1.89 3.44 1.41 0.93 1.18 1.83 1.06 1.43 1.02 1.12 2.49 1.19 1.96 0.71 1.70 1.30 1.08 1.03 1.57 2.00 0.77 1.11 Total 100.00 1001.00 100.00 100.00 100.00 100.00 99.99 99.98 99.98 100.00 1001.00 1001.00 100.00 1001.00 100.00 99.99 100.00 99.99 99.99 1001.00 100.00 99.99 100.00 100.00 100.00 S.I. 9.38 8.62 11.68 17.26 13.70 9.91 6.64 7.48 12.08 12.98 13.03 9.71 11.03 9.41 8.50 9.27 7.74 9.44 7.68 7.37 8.83 8.11 9.33 7.69 7.80 IEAOIA-HMCLALTERATI MINERALOGICAL-CHEMICAL Mg Val. 0.26 0.25 0.28 0.31 0.29 0.24 0.20 0.18 0.27 0.32 0.28 0.22 0.27 0.23 0.22 0.23 0.22 0.27 0.23 0.20 0.21 0.23 0.24 0.21 0.21 A.I. 0.81 0.81 0.76 0.83 0.71 0.71 0.77 0.72 0.71 0.86 0.71 0.73 0.76 0.69 0.79 0.78 0.82 0.81 0.80 0.76 0.74 0.70 0.71 0.75 0.75 Q 11.38 13.13 16.26 15.78 13.71 19.94 20.34 16.90 21.72 8.35 19.67 19.01 17.41 15.32 18.09 18.26 14.92 15.40 15.68 21.44 20.67 21.74 20.49 20.17 20.56 C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.39 0.00 0.00 7.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.40 0.00 0.00 0.00 Or 37.52 37.94 27.06 32.68 31.14 23.75 23.75 25.76 21.63 23.99 26.47 25.59 26.47 27.89 27.60 25.29 36.81 34.98 35.16 22.93 22.28 22.81 22.99 26.00 25.47 Ab 27.16 25.55 32.24 24.20 24.71 31.39 34.94 32.15 31.22 41.12 27.58 30.21 31.73 29.28 31.56 32.49 26.91 26.99 26.91 34.18 33.59 32.49 32.07 31.81 31.56 An 7.63 7.73 9.54 5.97 11.63 11.91 8.83 11.43 10.94 5.32 10.14 10.51 9.64 12.93 8.27 8.57 7.11 7.57 8.18 9.34 10.36 11.06 11.58 9.72 101.00 Di 1.30 1.14 0.80 1.28 0.81 2.00 0.76 0.31 0.48 4.94 0.00 0.40 0.73 0.00 0.85 0.64 0.98 0.99 0.71 0.44 0.36 0.00 3.00 0.14 0.18 Ed 2.97 2.83 1.65 2.49 1.57 7.00 2.45 1.13 1.07 8.68 0.00 1.20 1.61 0.00 2.47 1.85 2.77 2.25 1.94 1.37 1.07 0.00 7.00 0.42 0.54 Sl 4.27 3.97 2.45 3.77 2.37 9.00 3.21 1.44 1.56 13.62 0.00 1.61 2.34 0.00 3.32 2.49 3.74 3.24 2.65 1.81 1.42 0.00 9.00 0.56 0.71 En 1.86 1.69 2.39 3.76 2.86 2.11 1.09 1.50 2.29 1.02 2.89 1.93 2.20 2.12 1.55 1.77 1.51 1.91 1.56 1.37 1.70 1.69 1.96 1.63 1.61 Fs 4.85 4.81 5.63 8.38 6.38 6.39 4.06 6.26 5.82 2.06 7.07 6.58 5.58 6.62 5.19 5.86 4.92 4.94 4.92 4.92 5.86 5.30 5.79 5.81 5.67 Hy 6.72 6.49 8.02 12.15 9.25 8.50 5.15 7.76 8.12 3.08 9.96 8.51 7.79 8.73 6.74 7.64 6.44 6.85 6.48 6.29 7.56 7.00 7.74 7.44 7.29 Mt 0.95 0.92 0.95 1.33 1.06 0.94 0.81 1.00 0.92 0.98 1.04 1.06 0.95 0.96 0.94 0.97 0.93 0.89 0.87 0.85 0.94 0.79 0.86 0.89 0.88 Il 1.44 1.39 1.37 1.22 1.60 1.29 1.29 1.41 1.27 1.39 1.41 1.39 1.41 1.35 1.33 1.25 1.33 1.23 1.33 1.29 1.29 1.31 1.35 1.33 1.31 ONANDORIGINOFIGNIMBRITICSTONES Ap 1.34 1.07 0.49 0.44 0.60 0.35 0.37 0.51 0.42 0.63 1.44 0.63 0.69 0.53 0.56 0.65 0.65 0.72 1.02 0.39 0.42 0.44 0.46 0.88 0.69 D.I. 76.06 76.62 75.56 72.65 69.56 75.08 79.04 74.81 74.57 73.46 73.72 74.80 75.61 72.48 77.24 76.04 78.64 77.37 77.75 78.55 76.54 77.04 75.54 77.98 77.59 SAL 83.69 84.35 85.09 78.63 81.19 87.00 87.87 86.24 85.50 78.78 84.25 85.31 85.25 85.49 85.51 84.61 85.75 84.94 85.93 87.90 86.91 88.50 87.12 87.70 87.60 FEM 14.72 13.85 13.28 18.91 14.88 11.17 10.83 12.12 12.29 19.69 13.84 13.19 13.18 11.57 12.89 13.00 13.09 12.94 12.35 10.63 11.63 9.54 10.51 11.10 10.89 V 53.0 53.3 52.6 49.1 59.8 21.7 26.3 47.9 30.2 54.4 51.1 53.2 54.8 34.3 39.0 43.0 49.8 42.5 47.1 31.7 33.8 46.4 37.1 46.1 40.0 Ni 4.0 6.6 4.8 5.7 3.4 5.7 5.8 4.3 4.7 4.9 5.8 6.5 5.1 5.8 4.4 4.5 5.7 5.1 5.9 3.3 3.3 5.2 4.7 4.1 3.7 Zn 68.1 65.8 75.6 75.7 89.5 84.2 92.2 68.2 77.0 79.0 117.0 71.4 94.9 146.0 73.8 79.3 60.5 51.9 54.3 87.9 78.7 65.9 85.2 75.1 72.4 Rb 141 146 183 305 86.8 68.7 121 114 121 89.3 156 121 166 56.1 161 157 156 136 150 137 134 134 125 141 136 Sr 211 205 213 201 373 232 223 223 231 285 221 198 217 282 249 234 168 184 184 215 226 215 208 235 214 Ba 488 503 535 537 582 639 606 589 575 519 543 593 583 603 530 581 473 498 537 579 543 606 599 608 621 Zr 188 178 204 191 192 264 251 205 211 192 197 190 217 221 211 206 192 174 183 234 223 220 209 215 203 Pb n.d. 25.9 17.8 15.1 45.5 20.0 32.5 n.d. 22.1 20.9 32.9 n.d. 29.7 36.3 24.4 n.d. n.d. n.d. 13.6 18.3 12.5 15.9 23.8 20.6 n.d. Nb 12.6 9.9 10.6 10.9 10.5 14.8 13.2 12.3 10.9 5.7 12.5 8.2 9.6 12.8 11.8 12.1 10.5 8.7 7.4 12.0 9.9 13.0 8.3 14.3 10.8 Y 44.1 36.5 41.0 44.2 37.3 38.5 53.2 43.9 32.8 32.0 37.2 36.4 44.3 21.1 43.3 45.7 40.7 33.9 37.4 32.4 42.2 28.9 25.7 35.2 36.2 La 37.4 30.6 22.4 17.7 26.8 39.6 38.1 25.9 30.5 31.0 30.4 30.0 36.3 26.4 34.4 32.2 27.0 35.7 29.7 23.8 31.7 25.7 23.9 24.4 24.4 Ce 53.9 48.9 38.5 36.1 58.7 52.6 61.6 52.4 42.0 53.6 62.4 48.4 64.9 39.8 56.0 52.8 55.4 49.4 56.0 44.9 63.6 40.6 47.5 44.7 54.0

latite; QzTr = quartz-trachyte; Tr = trachyte; A.I. = agpaitic index of Shand (1951); D.I. = normative Quartz + Albite + Orthoclase + Nepheline + Kaliophilite + Leucite, diﬀerentiation index of Thornton and Tuttle (1961); n.d. = not detected. 411 412 Table 6. Chemical analysis of selected volcanics from Chilivani/Tula sector. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Sample CO18 CO41 CO107 CO182 CO215 CO109aS CO226 CO99 CO110 CO102 CO108b CO114 CO183 CO192 CO227 CO112 CO113c CO111 CO191 CO113b CO184 CO113a CO180 CO101a CO101b

− − − − − − − − − − − − − − − − − − − − − − − − − AL. ET COLUMBU S. Subsig. A A A A A A A A A A A A A A A A A A A A A A A A A Class. Ry qzTr qzTr qzTr qzTr qzTr qzTr qzTr qzTr qzLt qzTr Ry RyDc Ry RyDc Ry Ry Ry RyDc RyDc RyDc RyDc RyDc RyDc RyDc SiO2 65.85 66.81 67.20 62.85 66.49 67.05 65.37 66.57 66.96 67.91 66.47 72.81 67.04 65.30 71.03 69.35 69.08 72.59 69.63 70.34 71.49 69.01 70.97 70.33 70.10 TiO2 0.56 0.55 0.55 0.55 0.54 0.54 0.53 0.52 0.52 0.50 0.45 0.42 0.38 0.38 0.32 0.31 0.30 0.29 0.29 0.28 0.28 0.27 0.24 0.23 0.21 Al2O3 15.99 15.11 15.16 16.61 15.82 14.90 15.90 15.31 15.41 14.47 15.70 12.30 14.55 15.42 13.23 13.61 13.45 11.46 12.83 13.87 12.10 14.01 11.97 12.65 12.31 Fe2O3 4.35 3.83 3.79 4.10 3.56 3.40 4.33 3.38 3.68 3.74 2.85 2.81 3.38 3.00 2.21 2.15 2.64 1.97 2.42 2.52 2.31 1.77 2.44 1.52 1.62 FeO 0.12 0.56 0.32 0.14 0.36 0.69 0.22 0.70 0.43 0.35 0.58 0.44 0.33 0.35 0.34 0.57 1.38 0.70 0.21 0.90 0.43 0.81 0.27 0.82 0.50 MnO 7.00 7.00 0.10 6.00 6.00 8.00 7.00 5.00 0.11 0.10 0.10 3.00 7.00 8.00 0.13 8.00 4.00 0.10 6.00 4.00 5.00 4.00 4.00 4.00 7.00 MgO 0.55 0.29 0.38 1.03 0.30 0.49 0.40 0.31 0.29 0.33 0.17 0.23 1.70 0.89 0.94 1.00 1.31 0.70 1.39 1.43 1.05 1.23 1.20 0.96 1.20 CaO 1.40 1.44 1.49 0.60 1.36 1.32 1.47 0.99 1.39 1.62 0.64 1.07 2.92 1.54 1.70 1.58 1.61 1.65 1.97 2.00 2.11 1.67 2.16 2.01 1.96 Na2O 2.79 4.02 4.12 0.95 3.60 3.74 3.93 3.59 3.91 3.96 3.02 2.57 2.23 1.17 1.06 1.32 1.50 1.54 1.91 1.37 1.55 1.37 1.79 1.32 2.07 K2O 5.93 6.10 6.02 11.17 6.90 6.57 6.96 6.60 6.26 6.04 7.98 5.56 3.65 8.54 5.65 7.64 6.23 4.32 3.72 4.83 3.58 5.69 2.96 5.21 4.85 P2O5 0.13 0.13 0.13 0.12 0.16 0.13 0.12 5.00 9.00 0.20 0.10 0.19 6.00 6.00 6.00 8.00 4.00 5.00 7.00 6.00 0.15 6.00 5.00 4.00 0.11 L.O.I. 2.26 1.08 0.75 1.83 0.85 1.07 0.68 1.92 0.95 0.78 1.93 1.58 3.69 3.27 3.33 2.31 2.41 4.63 5.51 2.35 4.91 4.06 5.91 4.87 4.99 Total 100.00 99.99 1001.00 1001.00 100.00 99.98 99.98 99.99 100.00 100.00 99.99 1001.00 100.00 100.00 100.00 100.00 99.99 100.00 1001.00 99.99 1001.00 99.99 100.00 100.00 99.99 S.I. 5.93 2.79 3.61 7.83 2.78 4.54 3.54 2.95 2.77 3.19 1.52 2.75 22.43 8.40 12.29 104.00 14.49 10.67 19.80 18.74 16.99 14.84 20.17 12.82 14.78 Mg Val. 0.20 0.11 0.15 0.32 0.13 0.19 0.15 0.13 0.12 0.14 9.00 0.12 0.47 0.34 0.42 0.42 0.38 0.34 0.51 0.45 0.43 0.48 0.46 0.44 0.52 A.I. 0.69 0.87 0.88 0.82 0.85 0.89 0.88 0.85 0.86 0.90 0.87 0.83 0.52 0.72 0.59 0.77 0.68 0.63 0.56 0.54 0.53 0.60 0.51 0.62 0.70 Q 21.07 14.78 14.89 9.83 14.14 14.99 10.27 15.77 14.90 16.72 15.12 32.74 29.39 19.48 36.90 26.23 28.84 41.10 36.72 35.64 41.56 32.54 41.41 35.73 32.58 C 2.75 0.00 0.00 2.15 0.34 0.00 0.00 0.58 0.00 0.00 1.17 0.56 1.77 1.60 2.42 0.49 1.41 1.37 2.25 2.90 2.20 2.70 2.01 1.28 0.35 Or 35.04 36.05 35.57 66.00 40.77 38.82 41.13 39.00 36.99 35.69 47.15 32.85 21.57 50.46 33.39 45.15 36.81 25.53 21.98 28.54 21.15 33.62 17.49 30.79 28.66 Ab 23.61 34.01 34.86 8.04 30.46 31.64 33.25 30.38 33.08 33.51 25.55 21.74 18.87 9.90 8.97 11.17 12.69 13.03 16.16 11.59 13.11 11.59 15.15 11.17 17.51 An 6.10 5.17 5.09 2.19 5.70 4.46 5.19 4.58 6.01 3.87 2.52 4.07 14.09 7.25 8.04 7.32 7.73 7.86 9.32 9.53 9.49 7.89 10.39 9.71 9.00 Di 0.00 0.13 0.23 0.00 0.00 0.24 0.20 0.00 4.00 0.40 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ed 0.00 0.85 1.03 0.00 0.00 0.83 0.95 0.00 0.22 2.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sl 0.00 0.98 1.26 0.00 0.00 1.07 1.15 0.00 0.26 2.49 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 En 1.37 0.66 0.84 2.57 0.75 1.11 0.90 0.77 0.70 0.64 0.42 0.57 4.23 2.22 2.34 2.49 3.26 1.74 3.46 3.56 2.61 3.06 2.99 2.39 2.99 Fs 5.29 4.81 4.36 4.98 4.60 4.47 4.96 4.88 4.87 3.85 4.19 3.84 4.59 4.12 3.23 3.42 5.24 3.44 3.23 4.39 3.42 3.25 3.40 2.99 2.74 Hy 6.66 5.47 5.20 7.55 5.34 5.58 5.86 5.65 5.58 4.49 4.61 4.41 8.82 6.34 5.57 5.91 8.51 5.18 6.70 7.95 6.03 6.32 6.39 5.39 5.73 Mt 0.77 0.77 0.71 0.73 0.68 0.72 0.79 0.72 0.72 0.71 0.60 0.57 0.65 0.58 0.45 0.48 0.72 0.47 0.46 0.61 0.48 0.46 0.47 0.42 0.38 Il 1.06 1.04 1.04 1.04 1.03 1.03 1.01 0.99 0.99 0.95 0.85 0.80 0.72 0.72 0.61 0.59 0.57 0.55 0.55 0.53 0.53 0.51 0.46 0.44 0.40 Ap 0.30 0.30 0.30 0.28 0.37 0.30 0.28 0.12 0.21 0.46 0.23 0.44 0.14 0.14 0.14 0.19 9.00 0.12 0.16 0.14 0.35 0.14 0.12 9.00 0.25 D.I. 79.72 84.84 85.32 83.87 85.38 85.46 84.65 85.15 84.97 85.92 87.83 87.34 69.83 79.84 79.26 82.55 78.34 79.66 74.86 75.77 75.83 77.76 74.05 77.68 78.75 SAL 88.56 901.00 90.41 88.22 91.42 89.93 89.84 90.31 90.98 89.79 91.52 91.97 85.69 88.69 89.72 90.35 87.48 88.89 86.42 88.20 87.51 88.36 86.45 88.67 88.11 FEM 8.80 8.57 8.52 9.60 7.42 8.69 9.09 7.47 7.75 9.11 6.30 6.22 10.33 7.78 6.76 7.16 9.89 6.32 7.87 9.23 7.39 7.43 7.43 6.33 6.75 V 38.3 21.8 28.5 30.9 30.3 22.5 29.2 22.2 27.8 22.5 12.1 32.0 50.4 46.8 30.9 31.6 51.2 49.6 28.7 45.3 44.3 39.0 41.6 29.4 26.0 Ni 2.0 3.7 4.6 n.d. 5.3 3.4 6.0 6.7 5.7 4.1 11.0 4.8 2.4 7.1 6.0 6.2 7.7 4.9 4.2 6.7 2.8 4.6 5.4 4.6 5.0 Zn 43.3 69.0 73.5 28.9 69.5 71.4 70.9 61.7 68.0 57.6 30.5 21.9 42.7 51.7 41.4 49.2 57.7 37.2 43.1 45.4 31.6 44.1 37.5 40.5 32.8 Rb 176 252 257 204 273 279 266 296 254 244 282 168 121 244 201 219 273 84.3 132 144 84.9 154 179 158 117 Sr 138 192 193 49.7 187 172 177 152 187 180 113 137 155 345 203 187 285 487 1370 556 524 423 720 340 378 Ba 433 692 656 635 559 547 576 553 547 521 516 436 438 796 293 341 485 638 923 771 991 546 1130 557 479 Zr 170 291 288 159 263 287 271 294 284 263 301 184 162 214 210 203 110 86.7 159 99.9 66.9 105 90 108 96.6 Pb 15.0 29.5 15.9 19.8 19.5 26.4 29.2 n.d. 27.3 24.0 30.1 n.d. 19.7 23.2 n.d. n.d. 19.7 26.7 29.3 19.5 12.8 17.8 10.7 22.3 37.0 Nb 9.3 17.5 12.5 8.0 11.8 13.3 12.1 15.8 16.0 12.1 15.6 7.9 7.0 11.9 10.4 11.2 7.1 3.2 8.8 4.3 n.d. 3.8 4.6 4.1 4.9 Y 32.2 54.1 46.4 32.2 39.4 41.4 33.6 34.8 34.0 37.2 30.5 19.4 16.4 17.9 21.1 18.0 13.4 5.8 11.5 12.2 10.5 21.0 14.0 17.2 12.7 La 39.6 44.2 57.1 24.4 31.8 49.2 29.7 34.6 51.4 36.1 30.3 22.7 24.9 28.4 25.1 30.0 20.9 21.9 19.7 36.6 19.0 19.6 29.8 22.7 21.0 Ce 75.9 69.5 91.9 47.2 72.2 76.8 56.9 53.4 75.8 67.5 60.2 42.8 37.5 28.2 44.8 59.2 26.0 30.0 36.7 32.9 34.0 46.7 46.5 42.4 33.7

Major element data expressed in wt%, trace element data expressed in ppm. CIPW norm calculated setting Fe2O3/FeO wt% = 0.15. Abbreviations as in Table 5. MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 413

Figure 9. D.I. versus TiO2 diagram for the Oschiri-Chilivani/Tula sectors (only A− unaltered samples) with distinct sample populations from the LFW, IFC, MFE areas from Oschiri sector, and WPC and UPC areas from Chilivani/Tula sector. Graphic symbols: crosses = Castro Church samples; empty symbols = Otti Church samples; black filled-in symbols = outcrop samples of Oschiri and Chilivani/Tula sectors. The symbols of two churches and outcrop samples are differentiated according to the classification R1 vs. R2 diagram of De La Roche et al. (1980). Abbreviations: Tr = trachyte; qzTr = quartz-trachyte; RyDc = rhyodacite; qzLt = quartz-latite; Ry = rhyolite. Data from this work, Macciotta et al. (2001) and Columbu (2017). presence of typical structures (according to Lofgren (1971), sometimes represented by spherulites made up of radially-arranged acicular crystals. Due to incipient oxidation, in the more altered samples of the Church of Nostra Signora di Castro, as well as in LFW, IFC and MFE samples, the micropheno- Figure 10. Scanning electron microscope images (backscat- crystals of opaque granules have a cribriform structure, tered electrons) of the samples coming from the Castro Church, showing the devitrified volcanic ash clasts in the with consequent formation of secondary hematite and ignimbrite. precipitation of iron hydroxides. By SEM analysis, the groundmass glass is confirmed + to be largely devitrified (Figure 10). EDX microanalysis By SEM analysis of the specimens A taken from the shows that the cryptocrystalline assemblage mainly Church of Nostra Signora di Castro, abundant whitish consists of K-feldspar and quartz. Occasionally, milli- encrustations are observed on the exposed surface of metric lithophysaes are also present, often filled by sec- the ashlars (Figure 11), possibly induced by a combi- ondary K-feldspars and Fe-rich phyllosilicates. Within nation of epilithic lichens and moss. the intra-matrix porosity of the volcanic lithofacies As highligthed by EDX compositional spectrum with greenish chromatic features, there are fine- (Figure 11), the encrustations mainly constist of Ca- grained mica-like minerals, and occasionally micro- rich phase, possibly Ca-oxalates produced by lichens crystalline calcite is also present. and moss. XRD analyses, carried out on samples from the Church of Nostra Signora di Castro and from the out- Chemical data interpretation crops, indicate that the primary mineralogy consists of A total of 32 samples were interpreted considering the plagioclases, K-feldspars, quartz, clinopyroxene, ortho- chemical composition of the A− (‘fresh’ rock) and A+ pyroxene, biotite, and the presence of hematite as a (surface rock) portions of the material. Due to chemical secondary mineral phase. According to features changes linked to sporadic events (e.g. anthropic observed in thin section, there is also the presence activity), some samples were not considered in the dia- of subordinate phyllosilicates. XRD analysis on <2 μm grams. Some trace elements (i.e. Cr and Pb) were oriented mounts shows the presence of smectite excluded from all subsequent consideration because and mica-like minerals (i.e. celadonite and glauconite) the contents are below the detection limit (or non-cal- in the greenish samples, characterising the patches culable dispersion). with a more intensive greenish colour. Instead, these The analytical values (A−,A+) of major and trace minerals have not been found in the reddish elements were reported in the binary diagrams of samples. Moreover, the occasional presence of cristo- Figures 12 and 13, where the linear straight continuous balite, tridymite, and zeolite (i.e. mordenite) was line (with y = x and R2 = 1) and linear regression line of found. the samples (with an intercept equal to zero) are 414 S. COLUMBU ET AL.

Figure 11. Scanning electron microscope images of the surface of the stone ashlars from the Castro Church, which was exposed to atmospheric agents of deterioration. A, B: secondary (A) and back-scattered electrons (B) images of the same fragment, showing the presence of encrustations on the silicatic substrate; C: back-scattered electrons image of a Ca-bearing mineral neoformation (Ca- oxalate, Ca-carbonate?) on the silicatic substrate; D: Ca-bearing mineral neoformation with structures suggesting biological features. reported. Three main theoretical geochemical beha- and, subordinately, for Rb, Sr, and Ba and a clear viours can have occurred during the alteration increase for Ni, Y, and Ce. process: (1) the element does not vary (with sample The behaviour of the various chemical elements is points along the straight bisecting line passing through summarised analytically in Tables 7 and 8, where it is the origin of the axes); (2) the element decreases in possible to compare the regression equations and the outer portion A+ (with points under the bisecting some statistical parameters (e.g. correlation coeffi- line); or (3) the element increases in the outer portion cients, minimum, maximum, mean, standard deviation) A+ (with points above the bisecting line). To get useful of the two data sets (A− and A+) for the major (Table 7) graphical information from these diagrams on chemical and trace (Table 8) elements. modifications, it would be enough to observe the imbal- ances of the sample population with respect to the bisecting line, or the position of the regression dashed Discussion of results line (under or upper to the straight bisecting line in Geochemical features and provenance of church the diagrams) passing in the origin of axes. stones Considering the major elements (Figure 12), a decrease of concentration values in surface samples To obtain a definitive geochemical-petrographic + (A ) is observed for SiO2, TiO2,Al2O3,Na2O, and K2O, characterisation and classification of the hewn while a clear increase is observed for FeO (excluding ashlars, the macroscopic observations were compared an outlier), MnO, CaO, P2O5, L.O.I., and MgO concen- and integrated with microscopic and geochemical trations do not vary during the alteration process. ones. According to the diagram of De La Roche et al. In the case of trace elements, a decrease of concen- (1980) (modified; Figure 7), used as main classification tration values in outer samples (A+) is observed for V scheme of these pyroclastic rocks, the samples of the MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 415

Figure 12. Binary diagrams where reported A+ vs A− concentrations of major elements in samples from the Church. Solid line: x = y; dotted line: regression line of samples with intercept at the origin of axes x, y.

Church of Nostra Signora di Castro are mainly classified The microscopy analysis and the geochemical data as quartz-latite, rhyodacite, dacite, and rhyolite. That is show that the stone ashlars used to build the Church in agreement with the petrographic features and, also, of Nostra Signora di Castro have more similar petro- partly with the classification diagrams of Winchester graphic and geochemical characteristics to the volca- and Floyd (Figure 8) using some incompatible nics from Oschiri outcrops than those from the elements, Nb, Y, Zr, that indicate the presence of ande- Chilivani/Tula sector. sitic rock. The most frequently used pyroclastic facies for the Considering the alteration of these pyroclastics, it is ashlars of Castro monument show a petrographic probable that the anomalous andesitic composition of affinity with the volcanics from the western Oschiri the samples in these classification diagrams is due to subsector (LFW area) and also with a part of pyro- leaching of SiO2 (see Figure 12). Moreover, probably clastites used for the nearby Santa Maria di Otti the classification of some diagrams of Winchester Church (constructed in the same medieval period and Floyd (1977)(Figure 8(a, b)) was also affected by and similar on the basis of its architectural character- the increase of Y (see Figure 12), which affects a istics). This volcanic facies is characterised by the correct interpretation of chemical analysis, moving presence of yellow-ochre pumice lapilli with evident the classification of these rocks to more basic salic phenocrysts and by rare lithics (usually composition. rounded). In order to define the provenance areas of the hewn Only a subordinate amount of ashlars of the Church stones of monuments, a comparison between the of Nostra Signora di Castro resemble volcanics from the analytical data of samples from the Castro Church, central (IFC) and eastern (MFE) areas of Oschiri sector. the Otti Church, and volcanic outcrops of the field These latter rocks are characterised by parataxitic to sectors was made. massy vitreous structures, the presence of pumices 416 S. COLUMBU ET AL.

Figure 12. Continued

Table 7. Comparison of regression line equations (y = ax + b) and correlation coefficient data for 32 samples of the Castro Cathedral, using 64 chemical elements concentration data for each sample (#32 from A− inner portion and #32 from A+ outer portion) reported in Figures 12 and 13. Var. xyEquation (*) R2 (*) Equation R2 − + SiO2 A A y = 0.9928x 0.45185 y = 0.7237x + 16.911 0.52444 − + TiO2 A A y = 0.9777x −0.2117 y = 0.41147x + 0.402 0.24691 − + Al2O3 A A y = 0.9889x 0.4776 y = 0.7099x + 4.0144 0.56518 FeO A− A+ y = 1.0665x 0.00986 y = 0.6017x + 0.2499 0.20292 MnO A− A+ y = 1.1346x 0.55374 y = 1.1631x − 0.003 0.55409 MgO A− A+ y = 0.9834x 0.21147 y = 0.685x + 0.4111 0.26397 CaO A− A+ y = 1.07448x −0.1493 y = 0.4202x + 2.3362 0.09875 − + Na2OA A y = 0.9464x 0.20259 y = 0.8012x + 0.5545 0.2095 − + K2OAA y = 0.9843x 0.73032 y = 0.7656x + 1.1057 0.79724 − + P2O5 A A y = 1.1972x 0.18226 y = 0.8192x + 0.2257 0.27124 L.o.I. A− A+ y = 1.1428x 0.30181 y = 0.7808x + 1.119 0.40457 VA− A+ y = 0.8427x −0.0209 y = 0.2037x + 40.164 0.002236 Ni A− A+ y = 0.973x −1.2896 y = −0.1213x + 6.3058 0.0154 Zn A− A+ y = 0.863x −1.1936 y = 0.2669x + 58.693 0.37758 Rb A− A+ y = 0.9859x −0.0291 y = 0.4831x + 84.729 0.2194 Sr A− A+ y = 0.9695x 0.17108 y = 0.6058x + 83.819 0.26884 Ba A− A+ y = 0.8847x −6.7179 y = 0.0437x + 517.64 0.01731 Zr A− A+ y = 0.994x 0.45143 y = 0.7565x + 45.759 0.50099 Nb A− A+ y = 0.9999x 0.7941 y = −0.1075x + 11.99 0.00738 YA− A+ y = 1.04x 0.0388 y = 0.5684x + 18.724 0.12802 La A− A+ y = 0.9607x 0.3807 y = 0.1681x + 25.477 0.017737 Ce A− A+ y = 0.9268x −6.2115 y = −0.0637x + 61.394 0.02277 (*) with intercept equal to zero. *Data with intercept equal to zero, passing at the origin of axes x, y. Abbreviations: R2 = goodness of fit; R = correlation coefficients. MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 417

Table 8. Comparison of statistical parameters (minimum, maximum, mean, standard deviation) for 32 samples of the Castro Cathedral, using the chemical major and trace element concentration data (#32 from A− inner portion and #32 from A+ outer portion) reported in Figures 12 and 13. Var. x Min Max Mean St. Dev. y Min Max Mean St. Dev. − + SiO2 A 57.74 66.77 62.77 2.18 A 56.96 65.61 62.34 2.18 − + TiO2 A 0.55 0.84 0.71 0.06 A 0.50 0.79 0.70 0.05 − + Al2O3 A 12.43 16.10 14.34 0.89 A 12.40 16.43 14.19 0.84 FeO A− 0.09 1.50 0.34 0.26 A+ 0.10 1.51 0.45 0.35 MnO A− 0.05 0.14 0.10 0.02 A+ 0.04 0.20 0.11 0.04 MgO A− 0.82 1.84 1.31 0.29 A+ 0.60 2.26 1.31 0.39 CaO A− 2.17 4.93 3.45 0.65 A+ 2.34 6.23 3.79 0.88 − + Na2OA 3.35 4.52 3.80 0.28 A 1.69 4.47 3.60 0.50 − + K2OA 3.68 6.92 4.91 0.85 A 3.79 6.85 4.87 0.73 − + P2O5 A 0.17 1.51 0.39 0.29 A 0.21 1.96 0.54 0.46 L.o.I. A− 0.78 5.15 2.62 1.13 A+ 1.29 6.82 3.16 1.39 VA− 33.10 94.10 53.65 11.84 A+ 25.20 69.60 52.95 9.15 Ni A− 3.30 7.90 5.53 1.14 A+ 3.20 8.30 5.63 1.12 Zn A− 61.90 218.30 90.59 34.99 A+ 62.70 143.10 85.59 16.90 Rb A− 108.40 241.60 161.11 35.15 A+ 109.60 252.60 162.57 36.25 Sr A− 181.50 278.80 228.48 21.46 A+ 168.10 269.30 222.23 25.07 Ba A− 435.50 1013.50 585.88 133.80 A+ 476.60 700.60 543.25 44.44 Zr A− 174.20 220.00 191.98 11.73 A+ 171.50 226.10 190.99 12.53 Nb A− 6.80 14.20 10.58 1.65 A+ 6.00 14.40 10.85 2.06 YA− 29.90 47.90 39.22 4.43 A+ 24.20 54.30 41.02 7.04 La A− 20.40 45.60 31.18 5.55 A+ 17.90 47.80 30.72 7.08 Ce A− 32.90 159.90 54.67 20.32 A+ 43.20 76.40 57.91 8.58 varying in colour from green (Figure 4) to yellow-green. Chemical-mineralogical transformations and The latter facies have a greater abundance of sharp- physical decay edged (only slightly rounded) lithics with most In general, the ignimbritic rocks of Church of Nostra common size ∼ 5–6 mm. Phenocrystals of sialic min- Signora di Castro show a decay degree that mainly erals (Figure 4) are clearly visible to the naked eye. derives from sin- and epi-genetic alteration before The visible mafic crystals have always very small size the extraction of rocks (e.g. halmyrolysis processes) (<1 mm) and they are rare (or absent). (Ghiara et al. 1997; Cappelletti, Langella, and Cruciani Using as discriminating markers the differentiation 1999; Morbidelli et al. 1999; de Gennaro et al. 2000; index (D.I.) and TiO data (plotted in the diagram of 2 Cerri and Oggiano 2002; Steindlberger 2004). In agree- Figure 9), the samples belong mainly from the ment with Columbu et al. (2014), the results of XRD western area (LFW), subordinately from the central analysis show the presence of smectite and mica-like area (IFC) and rarely from the eastern area (MFE). Differ- minerals, as well as celadonite and glauconite, which ently from the case of Church of Nostra Signora di characterise the green patches. SEM-EDX analysis Castro, according to the results of the stepwise linear shows the presence of devitrified glasses in a ground- discriminant analysis and canonical analysis (Columbu mass of rock and the formation of phyllosilicates 2017) elaborated on the basis of 55 chemical analyses − with variable chemical composition, confirming the of geological samples (A type), the similar volcanics halmyrolytic alteration in the volcanic deposit during of Santa Maria di Otti Church derive mainly from the the sin-epigenetic processes. eastern area (MFE, closer to the monument) of Oschiri Occasional zeolite, tridymite, and cristobalite were sector and subordinately from the western area found. The zeolite is only present as mordenite, while (LFW). Only one sample comes from the IFC central the clinoptilolite is absent, contrary to what has been area. observed by other authors (Ghiara et al. 1997; Cappel- Thus, considering that on the hill immediately to the letti, Langella, and Cruciani 1999; dè Gennaro et al. west of the Church of Nostra Signora di Castro, but 2000) in other sectors of Sardinia. within the Chilivani/Tula sector, pyroclastic rocks However, processes that occurred in the monument outcrop having valid features as construction materials after the laying of minerals also affected the volcanic and easier to quarry due to their orientation, probably rocks from a chemical-mineralogical point of view. the ancient territorial subdivisions at those time (based These processes are mainly concentrated on the on property or administrative boundaries) affected the stone surface of monuments, with chemical changes choice of supply sites, regardless of distance. It can, between rock substrate (A−) and surface (A+), and therefore, be assumed that the choice of geomaterials with occasional formation of depositions with very was linked mainly to the questions of territorial jurisdic- small amounts of different secondary phases (e.g. tion between the Dioceses of San Antioco di Bisarcio, gypsum, Ca-carbonate). located further west, and the Curatoria of Nostra Considering the chemical data of both A− and A+ Signora di Castro (Figure 4). portions of the material taken from the Church of 418 S. COLUMBU ET AL.

Nostra Signora di Castro (Table 2), in agreement with rocks. Moreover, the metabolic activity of the same bio- results of Columbu (2017), SiO2,Na2O, and K2O deteriogens could inﬂuence the chemical transform- elements decrease during the alteration (Figure 12), ations of the rock surface, contributing to the showing a leaching process of the matrix on the smoothing process of some chemical elements in the surface of the stone. altered external portion (A+) of the samples.

In contrast, data relative to CaO, FeOT,P2O5, and L.O.I. (Figure 12) show greater enrichment in the A+ − Conclusions fraction as compared to the A one. On the basis of XRD analyses, CaO enrichment cannot be explained The stones of the old Nostra Signora di Castro Cathe- by the habit of whitewashing the walls of churches dral (Figures 2 and 3) are ignimbrite volcanics belong- with milk of lime. Analyses carried out on samples ing to the high-K calcalkaline (to shoshonitic) series of covered with organic material (i.e. guano) taken from Miocene volcanic deposits of Eastern Logudoro the outer walls support the hypothesis that high P2O5 (Figure 1) and outcropping in the Chilivani-Berchidda values derive from that source. Basin. The increase of L.O.I. determined at 1000°C high- The results highlight that in the Church of Nostra lights a clear alteration at the surface of the volcanic Signora di Castro several pyroclastic facies with rocks, due to a great weight loss for dehydration different petro-volcanological features (i.e. from low (with T < 400°C) and dehydroxylation (>400°C) of the to high welding degree, from eutaxitic to parataxitic hygroscopic minerals (i.e. smectite and phyllosilicate to vitreous structures, variable presence of pumice group, zeolite) more present in the outer A+ portion and lithics, from greenish to grey to pinkish to of stone with respect to the inner A portion. reddish colours) were used for carving the ashlars of + The great enrichment in FeOT in A is related to the the monument. According to the petrographic features weathering that mobilises iron by destabilising Fe- observed in thin section and the classification scheme bearing minerals, with the precipitation of hematite of De La Roche et al. (1980), the analysed samples of and stable Fe-hydroxides and the contemporaneous the Castro Cathedral can be classified as quartz-latite, mobilisation of other elements. rhyodacite, dacite and rhyolite and, subordinately, as In a chemico-physical point of view, the decay of the quartz-trachyte, trachyte, and trachyandesite. ignimbritic rocks from the Church of Nostra Signora di On a chemical and petrographic basis, the pyroclas- Castro, similarly to the case of the nearby church of tic rocks used in the Church of Nostra Signora di Castro Santa Maria di Otti (Columbu et al. 2014), is due to come from the Oschiri sector. According to the differ- two main coexisting processes: (1) precipitation and entiation index vs. TiO2 diagram, most of the samples cyclic hydration/dehydration of secondary hygroscopic belong to the western area, where the Less Fractio- phases (i.e. phyllosilicate, gypsum, zeolite, and soluble nated pyroclastites crop out (LFW), characterised by a salts coming from the soil and/or mortars); and (2) differentiation index range of 70–78, larger sialic phe- thermal and hydric dilatation, as a function of expo- nocrystals compared to mafic minerals, the presence sition to weathering. These processes are concentrated of yellow-ochre pumice, and rarely rounded lithics. especially at the base or in proximity to the covering of A subordinate part of the monument volcanic rocks the Church of Nostra Signora di Castro, where there is comes from the central Oschiri area, where the Interne- circulation of water for capillary rise or percolation, diately Fractionated pyroclastites crop out (IFC, with respectively. The decay in these zones leads to a pro- D.I. = 76–79), and only a few samples from eastern gressive loss of material of the façade with decohesion, Oschiri area, where the More Fractionated pyroclastic exfoliation, and flaking, which determine a retraction of rocks crop out (MFE), characterised by a differentation the outer surface of the ashlars (Figure 2) and a pro- index of 77–82. gressive obliteration of the decorative elements. Thus, the procurement of materials occurred from In agreement with various authors (e.g. Salvadori several outcrops in which different volcanic facies and Municchia (2016), the presence of biodeteriogens emerge. This also happened at the nearby Church of (especially on the north facades of the Church of Otti, where however, contrary to the case of the Nostra Signora di Castro) affects the decay of the Church of Nostra Signora di Castro, most of the vulca- ignimbritic rocks accelerating the chemical–physical nites come from the eastern area (MFE) of the Oschiri processes, although according to Özvan et al. (2015), sector, and in a subordinate way from the LFW in some cases the biological films have instead a pro- western area, and only rarely from the IFC central tective action of the stone against alteration processes. area (Columbu 2017). The lichens/moss produce encrustations mainly Considering that on the basis of geochemical and consisting of probable Ca-oxalates coming from their petrographic results in both churches there is no corre- metabolic activity that fix them on the stone surface. lation between the distribution of these volcanic facies In fact, according to Salvadori and Municchia (2016), (fortuitously employed in the monument) and their the oxalates can form on both carbonatic and silicate geographic origin, it seems unlikely that there could MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 419 be more quarries in operation simultaneously in the in that medieval epoch, and on the other, they allow to medieval period or different construction times for understand how the processes of alteration evolve in the utilisation of material. More probably, the choice this type of volcanic rock. Thus, the results obtained of materials was not based on technical aspects represent a significant technical-scientific base to be related to intrinsic properties, but on factors related taken into account during conservation procedures to the economic-social and political structure of that carried out on the monument, and also offer precise precise historical moment (i.e. contracts between indications where to procure the raw materials for private and ecclesiastical structures-diocese, workers, the possible substitutions of the most degraded taxes), where the ownership of land and outcrops stone ashlars, using pyroclastic rocks with physical- and especially territorial and administrative boundaries mechanical characteristics similar to those of the orig- were probably extremely important. The economic- inal materials. logistical aspects (e.g. road conditions, costs of trans- port and extraction) seem to have had a subordinate influence. Acknowledgments The pyroclastic rocks of Nostra Signora di Castro Sampling was carried out with the permission of the local Cathedral show evident alteration, both chemical and Superintendency to the Historical-Architectural Heritage of physical. The first starts with the halmyrolytic alteration Sassari and province. that occurred in the volcanic deposit during the singe- netic and epigenetic process. The results of SEM-EDX Disclosure statement and XRD analyses show an evident devitrification fl process of glass in groundmass of rock and the for- No potential con ict of interest was reported by the authors. mation of phyllosilicates with variable chemical composition, smectite, celadonite, and glauconite, and Funding occasional zeolite (i.e. mordenite). The research was supported financially by the funds of the Then, the chemical and mineralogical transform- Cagliari University (PRID 2015) and by the Regione Autonoma ations continued even after the laying of stones in della Sardegna (L.R. n.7 – 7 August 2007) through the the Church of Nostra Signora di Castro. The comparison research project entitled: ‘Romanesque and territory: the con- of concentration data of major and trace elements struction materials of the churches of the Judicial Sardinia: determined in the surface portion (A+) and the inner new approaches for the enhancement, conservation and res- ’ one (A−) of samples taken from the Castro Cathedral toration (application code CRP-18095, CUP: F71J11000620002). highlights a leaching process in pyroclastite samples.

SiO2,Al2O3,Na2O, and K2O decrease during the alteration process, due to the hydrolysis mainly of silicate ORCID phases (feldspars) and subordinately of a glassy Anna Gioncada http://orcid.org/0000-0002-8513-7377 matrix of rock. Greater enrichment of CaO, FeOt,P2O5, and L.O.I. on the surface of the rock is also observed. The CaO increase in A+ is due to the re-precipitation References of Ca-carbonates on the stone surface, resulting from Advokaat, E. L., D. J. J. Van Hinsbergen, M. Maffione, C. G. binder leaching/dissolution of ancient based-lime Langereis, R. L. M. Vissers, A. Cherchi, R. Schroeder, H. bedding mortars or (more probably) treatments used Madani, and S. Columbu. 2014. “Eocene Rotation of for preservation or colouring decorations of stone Sardinia, and the Paleogeography of the Western facades. The highest value of phosphorous oxide in Mediterranean Region.” Earth and Planetary Science – the A+ rock portion is due to the nitrates contained in Letters 401: 183 195. Antonelli, F., S. Columbu, M. De Vos Raaijmakers, and M. the guano of animals (e.g. birds). L.O.I. increase Andreoli. 2014. “An Archaeometric Contribution to the confirms an advanced alteration degree in the stone Study of Ancient Millstones from the Mulargia Area surface. (Sardinia, Italy) Through New Analytical Data on Volcanic Moreover, the presence of biodeteriogens (lichens, raw Material and Archaeological Items from Hellenistic ” mosses, etc.) in the facades addressed to the north con- and Roman North Africa. Journal of Archaeological Science 50: 243–261. tributes to the deterioration of the church’s ignimbrites, Antonelli, F., S. Columbu, M. Lezzerini, and D. Miriello. 2014. through the production of secondary compounds on “Petrographic Characterization and Provenance the stone surface based mainly on Ca-oxalates, and Determination of the White Marbles Used in the Roman through a physical action due to the constant presence Sculptures of Forum Sempronii (Fossombrone, Marche, of moisture. Italy).” Applied Physics A 115: 1033–1040. “ The investigations carried out in this research on the Bargossi, G. M., P. Felli, and F. Guerriri. 2002. Pietra serena. Materia della città.” Aida ed., Firenze, pp. 200. pyroclastics of the Nostra Signora di Castro Church Beccaluva, L., G. Bianchini, C. Bonadiman, M. Coltorti, G. have highlighted, on the one hand, identification of Macciotta, F. Siena, and C. Vaccaro. 2005. “Within-plate the geological outcrops used for the building materials Cenozoic Volcanism and Lithospheric Mantle Evolution in 420 S. COLUMBU ET AL.

the Western-central Mediterranean Area.” In Elsevier Special Geochemical Statistical Approach.” Applied Physics A Volume, “Crop Project — Deep Seismic Exploration of the Materials Science and Processing 123 (3): 1–28. Central Mediterranean and Italy”, edited by I. Finetti, 641– Columbu, S., F. Antonelli, and F. Sitzia. 2018. “Origin of Roman 664. Amsterdam: Elsevier. Worked Stones From St. Saturno Christian Basilica (South Beccaluva, L., G. Bianchini, M. Coltorti, F. Siena, and M. Verde. Sardinia, Italy).” Mediterranean Archaeology and 2005. “Cenozoic Tectono- Magmatic Evolution of the Archaeometry 18 (5): 17–36. Central-Western Mediterranean: Migration of an Arc- Columbu, S., G. Cruciani, D. Fancello, M. Franceschelli, and G. Interarc Basin System and Variations in the Mode of Musumeci. 2015. “Petrophysical Properties of a Granite- Subduction.” In Elsevier Special Volume, “Crop Project— Protomylonite-Ultramylonite Sequence: Insight From the Deep Seismic Exploration of the Central Mediterranean and Monte Grighini Shear Zone, Central Sardinia, Italy.” Italy”, edited by I. Finetti, 623–640. Amsterdam: Elsevier. European Journal of Mineralogy 27 (4): 471–486. Beccaluva, L., G. Bianchini, C. Natali, and F. Siena. 2011. Columbu, S., and A. M. Garau. 2017. “Mineralogical, “Geodynamic Control on Orogenic and Anorogenic Petrographic and Chemical Analysis of Geomaterials Magmatic Phases in Sardinia and Southern Spain: Used in the Mortars of Roman Nora Theatre (South Inferences for the Cenozoic Evolution of the Western Sardinia, Italy).” Italian Journal of Geosciences 136: 238–262. Mediterranean.” Lithos 123 (1–4): 218–224. Columbu, S., A. M. Garau, and C. Lugliè. 2018. “Geochemical Beccaluva, L., P. Brotzu, G. Macciotta, L. Morbidelli, G. Serri, Characterisation of Pozzolanic Obsidian Glasses Used in and G. Traversa. 1989. “Cainozoic Tectono-Magmatic the Ancient Mortars of Nora Roman Theatre (Sardinia, Evolution and Inferred Mantle Sources in the Sardo- Italy): Provenance of Raw Materials and Historical– Tyrrenian Area.” In The Lithosphere in Italy, edited by A. Archaeological Implications.” Archaeological and Boriani, M. Bonafede, G. B. Piccardo, and G. B. Vai, pp. Anthropological Sciences. Advance online publication. 229–248. Roma. Advances in Earth Science Research. Atti doi:10.1007/s12520-018-0658-y. Conv. Acc. Naz. Lincei, No. 80. Columbu, S., A. Gioncada, M. Lezzerini, and M. Marchi. 2014. Beccaluva, L., L. Civetta, G. Macciotta, and C. A. Ricci. 1985. “Hydric Dilatation of Ignimbritic Stones Used in the “Geochronology in Sardinia: Results and Problems.” Church of Santa Maria di Otti (Oschiri, Northern Sardinia, Rendiconti della Società Ialiana. di Mineralogia e Petrologia Italy).” Italian Journal of Geosciences 133: 149–160. 40: 57–72. Columbu, S., C. Lisci, F. Sitzia, and G. Buccellato. 2017. Benavente, D., G. Cultrone, and M. Gómez-Heras. 2008. “The “Physical-mechanical Consolidation and Protection of Combined Influence of Mineralogical, Hygric and Thermal Miocenic Limestone Used on Mediterranean Historical Properties on the Durability of Porous Building Stones.” Monuments: The Case Study of Pietra Cantone European Journal of Mineralogy 20 (4): 673–685. (Southern Sardinia, Italy).” Environmental Earth Sciences Burrus, J. 1984. “Contribution to a Geodynamic Synthesis of 76 (4): 1–29. the Provençal Basin (North-Western Mediterranean).” Columbu, S., C. Lisci, F. Sitzia, G. Lorenzetti, M. Lezzerini, S. Marine Geology 55: 247–269. Pagnotta, S. Raneri, et al. 2018. “Mineralogical, Cappelletti, P., A. Langella, and G. Cruciani. 1999. “Crystal- Petrographic and Physical-mechanical Study of Roman chemistry and Synchrotron Rietveld Refinement of Two Construction Materials from the Maritime Theatre of Different Clinoptilolites From Volcanoclastites of North- Hadrian’s Villa (Rome, Italy).” Measurement 127: 264–276. Western Sardinia.” European Journal of Mineralogy 11 (6): doi:10.1016/j.measurement.2018.05.103. 1051–1060. Columbu, S., M. Palomba, F. Sitzia, and M. Murgia. 2018. Carmignani, L., S. Barca, G. Oggiano, I. Pertusati, P. Conti, A. “Geochemical, Mineral-Petrographic and Physical- Eltrudis, A. Funedda, and S. Pasci. 1996. “Carta Geologica Mechanical Characterization of Stones and Mortars From della Sardegna 1:200.000.” Serv Geol d’Italia. the Romanesque Saccargia Basilica (Sardinia, Italy) to Carminati, E., and C. Doglioni. 2012. “Alps vs. Apennines: The Define Their Origin and Alteration.” Italian Journal of Paradigm of a Tectonically Asymmetric Earth.” Earth- Geosciences 137: 369–695. Science Reviews 112: 67–96. Columbu, S., G. Piras, F. Sitzia, S. Pagnotta, S. Raneri, S. Cerri, G., and G. Oggiano. 2002. “Le epiclastiti zeolitizzate del Legnaioli, V. Palleschi, M. Lezzerini, and M. Giamello. Logudoro orientale: un livello guida all’interno della suc- 2018. “Petrographic and Mineralogical Characterization of cessione vulcano sedimentaria della Sardegna centro set- Volcanic Rocks and Surface-Depositions on Romanesque tentrionale.” Bollettino della Società Geologica Italiana 121 Monuments.” Mediterranean Archaeology and (1): 3–10. Archaeometry 18 (5): 37–63. Cherchi, A., N. Mancin, L. Montadert, M. Murru, M. T. Putzu, F. Columbu, S., F. Sitzia, and G. Ennas. 2017. “The Ancient Schiavinotto, and V. Verrubbi. 2008. “The Stratigraphic Pozzolanic Mortars and Concretes of Heliocaminus Baths Response to the Oligo-Miocene Extension in the Western in Hadrian’s Villa (Tivoli, Italy).” Archaeological and Mediterranean From Observations on the Sardinia Anthropological Sciences 9: 523–553. Graben System (Italy).” Bulletin de la Societe Geologique Columbu, S., F. Sitzia, and G. Verdiani. 2015. “Contribution of de France 179: 267–287. Petrophysical Analysis and 3D Digital Survey in the Cherchi, A., and L. Montadert. 1982. “Oligo-Miocene Rift of Archaeometric Investigations of the Emperor Hadrian’s Sardinia and the Early History of the Western Baths (Tivoli, Italy).” Rendiconti Lincei 26: 455–474. Mediterranean Basin.” Nature 298: 736–739. Columbu, S., and G. Verdiani. 2014. “Digital Survey and Chiarelli, N., D. Miriello, G. Bianchi, G. Fichera, M. Giamello, and Material Analysis Strategies for Documenting, Monitoring I. T. Memmi. 2015. “Characterisation of Ancient Mortars and Study the Romanesque Churches in Sardinia, Italy.” from the S. Niccolò Archaeological Complex in Montieri Lecture Notes in Computer Science, Springer 8740: 446–453. (Tuscany – Italy).” Construction and Building Materials 96: Coroneo, R. 1993. Architettura romanica dalla metà del Mille al 442–460. primo ‘300. Nuoro: Ilisso. Columbu, S. 2017. “Provenance and Alteration of Pyroclastic Coulon, C. 1977. “Le volcanism calco-alcalin cenozoique de Rocks From the Romanesque Churches of Logudoro Sardaigne, Italie.” Unpublished master’s thesis. Univesité (North Sardinia, Italy) Using a Petrographic and St. Jerome, Marseille. MINERALOGICAL-CHEMICAL ALTERATIONANDORIGINOFIGNIMBRITICSTONES 421

Cruciani, G., A. Dini, M. Franceschelli, M. Puxeddu, and D. Lezzerini, M., F. Antonelli, S. Columbu, R. Gadducci, A. Marradi, Utzeri. 2010. “Metabasite from the Variscan Belt in NE D. Miriello, L. Parodi, L. Secchiari, and A. Lazzeri. 2016. “The Sardinia, Italy: Within-Plate OIB-Like Melts with Very High Documentation and Conservation of the Cultural Sr and Low Nd Isotope Ratios.” European Journal of Heritage: 3D Laser Scanning and Gis Techniques for Mineralogy 22: 509–523. Thematic Mapping of the Stonework of the Façade dè Gennaro, M., D. Calcaterra, P. Cappelletti, A. Langella, and of St. Nicholas Church (Pisa, Italy).” International Journal V. Morra. 2000. “Building Stone and Related Weathering in of Architectural Heritage: Conservation, Analysis, and the Architecture of the Ancient City of Naples.” Journal of Restoration 10: 9–19. Cultural Heritage 1 (4): 399–414. Lezzerini, M., S. Pagnotta, S. Columbu, and G. Gallello. 2018. De La Roche, H., J. Leterrier, P. Grandclaude, and M. Marchal. “Archaeometric Study of Mortars from the Pisa’s 1980. “A Classification of Volcanic and Plutonic Rocks Using Cathedral Square (Italy).” Measurement 126: 322–331. R 1 R 2 -Diagram and Major-Element Analyses — Its Lofgren, G. 1971. “Experimentally Produced Devitrification Relationships with Current Nomenclature.” Chemical Textures in Natural Rhyolitic Glass.” Geological Society of Geology 29: 183–210. America Bulletin 82: 111–124. Di Benedetto, C., P. Cappelletti, M. Favaro, S. F. Graziano, A. López-Doncel, R., W. Wedekind, R. Dohrmann, and S. Langella, D. Calcaterra, and A. Colella. 2015. “Porosity as Siegesmund. 2013. “Moisture Expansion Associated to Key Factor in the Durability of Two Historical Building Secondary Porosity: An Example of the Loseros Tuff of Stones: Neapolitan Yellow Tuff and Vicenza Stone.” Guanajuato, Mexico.” Environmental Earth Sciences 69: Engineering Geology 193: 310–319. 1189–1201. Dostal, J., C. Coulon, and C. Dupuy. 1982. “Cainozoic Andesitic Lustrino, M., S. Duggen, and C. L. Rosenberg. 2011. “The Rock of Sardinia (Italy).” In Andesites: Orogenic Andesites Central-Western Mediterranean: Anomalous Igneous and Related Rocks, edited by R. S. Thorpe, 353–370. Activity in an Anomalous Collisional Tectonic Setting.” Chichester: J. Wiley & Sons. Earth-Science Reviews 104: 1–40. Franzini, M., L. Leoni, M. Lezzerini, and R. Cardelli. 2007. Lustrino, M., L. Fedele, L. Melluso, V. Morra, F. Ronga, J. “Relationships Between Mineralogical Composition, Water Geldmacher, D. Duggen, et al. 2013. “Origin and Absorption and Hydric Dilatation in the "Macigno" Evolution of Cenozoic Magmatism of Sardinia (Italy). A Sandstones From Lunigiana (Massa, Tuscany).” European Combined Isotopic (Sr–Nd–Pb–O–Hf–Os) and Petrological Journal of Mineralogy 19 (1): 113–123. View.” Lithos 180–181: 138–158. Franzini, M., L. Leoni, and M. Saitta. 1972. “A Simple Method to Lustrino,M.,V.Morra,L.Fedele,andL.Franciosi.2009. “Beginning Evaluate the Matrix Effects in X-Ray Fluorescence Analysis.” of the Apennine Subduction System in Central Western X-Ray Spectrometry 1: 151–154. Mediterranean: Constraints from Cenozoic ‘Orogenic’ Gattaceca, J., A. Deino, R. Rizzo, D. S. Jones, B. Henry, B. Magmatic Activity of Sardinia, Italy.” Tectonics 28: 1–23. Beaudoin, and F. Vadeboin. 2007. “Miocene Rotation of Manzoni, E., F. Casarin, S. Dandria, A. Dusi, M. Giamello, E. Sardinia: New Paleomagnetic and Geochronological Giorgi, F. Gabbrielli, A. Magrini, and F. Randazzo. 2019. Constraints and Geodynamic Implications.” Earth and “The Medieval Grange of Cuna – Siena (Italy) – Planetary Science Letters 258: 359–377. Interdisciplinary Studies on Masonry Structures.” RILEM Ghiara, M. R., R. Lonis, C. Petti, E. Franco, S. Luxoro, and G. Bookseries 18: 2350-2358. Balassone. 1997. “The Zeolitization Process of Tertiary Macciotta, G., G. Bertorino, A. M. Caredda, S. Columbu, R. Orogenic Ignimbrites From Sardinia (Italy): Distribution Coroneo, M. Franceschelli, M. Marchi, S. Rescic. 2001. The and Mining Importance.” Periodico Mineralogia 66: 211–229. S.Antioco of Bisarcio Basilica (NE Sardinia, Italy): Water- Gioncada, A., L. Leoni, M. Lezzerini, and D. Miriello. 2011. rock Interaction in Ignimbrite Monument Decay. In “Relationships Between Mineralogical and Textural Water-Rock Interaction (WRI-10), edited by R. Cidu, vol. 1, Factors in Respect to Hydric Dilatation of Some 415–418. Lisse: Swets & Zeitlinger. Sandstones and Meta-Sandstones from the Northern Miriello, D., F. Antonelli, C. Apollaro, A. Bloise, N. Bruno, E. Apennine.” Italian Journal of Geosciences 130 (3): 394–403. Catalano, S. Columbu, et al. 2015. “New Data About the Giraud, J., H. Bellon, and G. Turco. 1979.L’intrusion microdiori- Ancient Mortars from the Archaeological Site of Kyme tique tertiaire d’Alghero (Sardaigne). Age K/Ar et relation (Turkey): Compositional Characterization.” Periodico di avec le magmatism calco-alcalin sarde. Analogies aves les Mineralogia 84 (3A): 497–517. esterellites de l’Esterel (Var). C. R. Acad. Sci. Paris: 288, 9– Montigny, R., J. B. Edel, and R. Thuizat. 1981. “Oligo-miocene 12, Paris. Rotation of Sardinia: k-ar Ages and Paleomagnetic Data Irvine, T. N., and W. R. A. Baragar. 1971. “A Guide to the of Tertiary Volcanics.” Earth and Planetary Science Letters Chemical Classification of the Common Volcanic Rocks.” 54: 261–271. Canadian Journal of Earth Sciences 8: 523–548. Morbidelli, P., M. R. Ghiara, R. Lonis R, and A. Sau. 1999. Lecca, L., R. Lonis, S. Luxoro, E. Melis, F. Secchi, and P. Brotzu. “Zeolitic Occurrences From Tertiary Pyroclastic Flows and 1997. “Oligo–Miocene Volcanic Sequences and Rifting Related Epiclastic Deposits Outcropping in Northern Stages in Sardinia: A Review.” Periodico di Mineralogia 66: Sardinia (Italy).” Periodico Mineralogia 2: 287–313. 7–61. Oggiano, G., S. Pasci, and A. Funedda. 1995. “Il bacino di Le Maitre, R. W., A. Streckeisen, B. Zanettin, M. J. Le Bas, B. Chilivani- Berchidda: un esempio di struttura transtensiva. Bonin, P. Bateman, G. Bellieni, et al. 2002. Igneous Rocks: Possibili relazioni con la geodinamica cenozoica del A Classification and Glossary of Terms. Recommendations Mediterraneo occidentale.” Bollettino della Società of the International Union of Geological Sciences, Geologica Italiana 114: 465–475. Subcommission of the Systematics of Igneous Rocks. Özvan, A., I. Dinçer, M. Akın, V. Oyan, and Mu. Tapan. 2015. Cambridge: Cambridge University Press. “Experimental Studies on Ignimbrite and the Effect of Leoni, L., M. Lezzerini, S. Battaglia, and F. Cavalcante. 2010. Lichens and Capillarity on the Deterioration of Seljuk “Corrensite and Chlorite-Rich Chl-S Mixed Layers in Gravestones.” Engineering Geology 185: 81–95. Sandstones From the‘Macigno’ Formation (Northwestern Peterson, D. W. 1979. “Significance of Flattening of Pumice Tuscany, Italy).” Clay Minerals 45: 87–106. Fragments in Ash-flow Tuffs.” In Ash-Flow Tuffs and 422 S. COLUMBU ET AL.

Associate Igneous Rocks, edited by C.E. Chapin and Savelli, C., L. Beccaluva, M. Deriu, G. Macciotta, and L. W. E. Elston. Spec. Pap. Geol. Soc. Am. 180:95–204. Maccioni. 1979. “K/Ar Geochronology and Evolution of Pioli, L., and M. and Rosi. 2005. “Rheomorphic Structures in a the Tertiary “Calc-Alkalic” Volcanism of Sardinia (Italy).” High-Grade Ignimbrite: The Nuraxi Tuff, Sulcis Volcanic Journal of Volcanology and Geothermal Research 5: 257– District (SW Sardinia, Italy).” Journal of Volcanology and 269. Geothermal Research 142: 11–28. Sebastian, E., G. Cultrone, D. Benavente, L. L. Fernandez, K. Quane, S. L., and J. K. Russell. 2005a. “Ranking Welding Elert, and C. Rodriguez-Navarro. 2008. “Swelling Damage Intensity in Pyroclastic Deposits.” Bulletin of Volcanology, in Clay-Rich Sandstones Used in the Church of San 67 (2): 129–143. Mateo in Tarifa (Spain).” Journal of Cultural Heritage 9 Quane, S. L., and J. K. Russell. 2005b. “Welding: Insights from (2): 66–76. Hightemperature Analogue Experiments.” Journal of Serra, R., and R. Coroneo. 2004. Sardegna Preromanica e Volcanology and Geothermal Research 142: 67–87. Romanica. Milano: Jaca Book, pp. 320. Ragan, D. H., and M. F. Sheridan. 1972. “Compaction of the Shand, S. J. 1951. Eruptive Rocks; Their Genesis, Composition, Bishop Tuff, California.” Geological Society of America Classification, and Their Relation to Ore-Deposits, with a Bulletin 83: 95–106. Chapter on Meteorites. 4th ed. Wiley: New York. Ramacciotti, M., S. Rubio, G. Gallello, M. Lezzerini, S. Columbu, Smith, R. L. 1960. “Ash Flows.” Geological Society of America E. Hernandez, A. Morales-Rubio, A. Pastor, and M. De La Bulletin 71: 795–842. Guardia. 2018. “Chronological Classification of Ancient Spano, G. 1858. “Sardegna Sacra e le Antiche Diocesi.” Mortars Employing Spectroscopy and Spectrometry Bullettino Archeologico Sardo 4: 41–48. Techniques: Sagunto (Valencia, Spain) Case.” Journal of Sparks, R. S. J., S. R. Tait, and Y. Yanev. 1999. “Dense Welding Spectroscopy 2018: 1–10. doi:10.1155/2018/9736547. Caused by Volatile Resorption.” Journal of the Geological Raneri, S., S. Pagnotta, M. Lezzerini, S. Legnaioli, V. Palleschi, S. Society 156: 217–225. Columbu, N. F. Neri, and P. Mazzoleni. 2018. “Examining the Speranza, F., I. M. Villa, L. Sagnotti, F. Florindo, D. Cosentino, P. Reactivity of Volcanic Ash in Ancient Mortars by Using a Cipollari, and M. Mattei. 2002. “Age of the Corsica-Sardinia Micro-chemical Approach.” Mediterranean Archaeology Rotation and Liguro-Provençal Basin Spreading: New and Archaeometry 18 (5): 147–157. Paleomagnetic and Ar/Ar Evidence.” Tectonophysics 347 Riehle, J. R. 1973. “Calculated Compaction Profiles of Rhyolitic (4): 231–251. Ash-flow Tuffs: Computational Model.” Bulletin of Steindlberger, E. 2004. “Volcanic Tuffs From Hesse (Germany) Volcanology 57: 319–336. and Their Weathering Behaviour.” Environmental Geology Rodriguez-Navarro, C., E. Hansen, E. Sebastián, and W. S. 46: 378–390. Ginell. 1997. “The Role of Clays in the Decay of Ancient Streck, M. J., and A. L. Grunder. 1995. “Crystallization and Egyptian Limestone Sculptures.” Journal of the American Welding Variations in a Widespread Ignimbrite Sheet; the Institute for Conservation 36: 151–163. Rattlesnake Tuff, Eastern Oregon, USA.” Bulletin of Sabatini, G., M. Giamello, R. Pini, S. Siano, and R. Salimbeni. Volcanology 57: 151–169. 2000. “Laser Cleaning Methodologies for Stone Façades Thornton, C. P., and O. F. Tuttle. 1960. “Chemistry of Igneous and Monuments: Laboratory Analyses on Lithotypes of Rocks--[Part] 1, Differentiation Index.” American Journal of Siena Architecture.” Journal of Cultural Heritage 1 (1): S9– Science 258: 664–684. S19. Vardabasso, S., and A. Atzeni. 1962. “Il bacino Oligocenico di Salvadori, O., and A. C. Municchia. 2016. “The Role of Fungi Oschiri- Berchidda nella Sardegna nord-occidentale.” and Lichens in the Biodeterioration of Stone Memorie della Societa Geologica Italiana 3: 717. Monuments.” The Open Conference Proceedings Journal 7 Verdiani, G., and S. Columbu. 2010. “E. Stone, an Archive for (1: M4): 39–54. the Sardinia Monumental Witnesses.” Lecture Notes in Sau, A., L. Lecca, R. Lonis, F. Secchi, and M. L. Fercia. 2005. “La Computer Science 6436: 356–372. seconda fase del Rift Sardo: vulcanismo ed evoluzione dei Wangler, T., and G. W. Scherer. 2008. “Clay Swelling subbacini di Ardara-Chilivani e Bonorva (Sardegna setten- Mechanism in Clay-Bearing Sandstones.” Environmental trionale).” Bollettino della Società Geologica Italiana 124: Geology 56 (3–4): 529–534. 3–20. Winchester, J. A., and P. A. Floyd. 1977. “Geochemical Savelli, C. 1975. “Datazioni preliminari col metodo K-Ar di vul- Discrimination of Different Magma Series and Their caniti della Sardegna sud-occidentale.” Rendiconti della Differentiation Products Using Immobile Elements.” Societià Italiana di Mineralogia e Petrologia 31: 191–198. Chemical Geology 20: 325–343.